"ajm04","Tools for Scientific Discovery","ajm","Ashesh Mahidadia","","","Active","BINFBIOMCECSSE","R & D","","Recent rapid advances in biological sciences have resulted in massive amount of data and domain theories. The project aims to build an interactive tool that helps experimental scientists to quickly evaluate and interpret the available, possibly massive amount of, data. ","","","",""
"AJM08","Complementary Predictive Modelling","AJM","Ashesh Mahidadia","","","Active","BINFBIOMCECSSE","R & D","","The project investigates predictive modelling/machine learning/data mining techniques (regression and classification) that augment predictive scoring with novel risk and innovation measures. These measures present a richer set of supervised and unsupervised semantics on the nature of each prediction, allowing more considered decision making. Augmented models are more conservative and selective in their decision making, taking into account concept drift, small training set problems and heretofore undetected forms of model risk. The student will learn about commercially relevant methods in data mining and analytics, and  benefit from supervision by an industry leader in the field (Dr Eugene Dubossarsky).","","","",""
"AJT41","A Handheld System for Plant Identification","AJT","Andrew Taylor","","","Active","BINFBIOMCECSSE","R & D","","Development of a plant identification system which data-mines testual plant descriptions in to build a database which can be queried with simple natural language","","good programming skills ","",""
"AN01","Systematic code generation in compilation","AN","Albert Nymeyer","","","Active","BINFBIOMCECSSE","R & D","","Code generators are usually hacked together. In this research a method is developed that allows a code generator (for a compiler) to be generated automatically. There is an algorithm, and a prototype written in C. Both need to be developed further and benchmarking carried out,","The supervisor has written papers on the method that are available.","Programming skills, ability to puzzle, no knowledge of parsing required","A C compiler.",""
"AN05","Training smart search algorithms","AN","Albert Nymeyer","","","Active","BINFBIOMCECSSE","R & D","","Heuristics are educated guesses or costs how good a particular state of the system is. They are usually employed by search algorithms to improve performance. Heuristics are static however. As a search algorithm wanders through a search space looking for the best path to a goal state it gathers information that can be used to improve the quality of the heuristic, and find an even better path. This is training the heuristic as it is being used.","Any good AI book.","Some knowledge of AI and search techniques would help.","none",""
"AN06","Smart Search in Model Checking","AN","Albert Nymeyer","","","Active","BINFBIOMCECSSE","R & D","","Model checking is intrinsically a brute-force method that analyses the whole state space of a model in search of states that violate some property. The model is expressed as a state transition graph and the property as a temporal logic formula. This is required as a model checker has traditionally been used as a verification tool.In this work, the brute-force approach is replaced by a smarter, heuristically-driven search algorithm that uses domain knowledge to steer the model checker in the direction of violating states. The resulting model checker can be much faster than a conventional model checker in finding `bad' states, and as such, fulfils the role of a `debugger' rather than a verifier.The focus of the research is to determine what domain knowledge should be used, and how to represent it as a heuristic. In fact, the domain knowledge is just an abstract view of the original problem, and moreover, a `relaxed' version of the original system can be used as a smart heuristic.An implementation of a system in which a software engineer can `debug' his system by abstracting it and searching for particular bad states is currently being worked on.","","Some affinity with logic or concurrency is an advantage.","",""
"AN08","A Software Project Management Tool","AN","Albert Nymeyer","","","Active","BINFBIOMCECSSE","Development","","A very important component of the SENG workshops at CSE is project management. To do project management properly you need to have a tool. This tool would allow you to assign and schedule tasks, monitor progress, document the development, do configuration management, and generally work together as a team. Extra features would be a risk analysis procedure and integration with the B-toolkit.There are tools available that do these things and lots more, but they are usually large and difficult to master. In this project we seek a light-weight tool that is tailored to the needs of the workshops, and easy to use.The work requires an individual or group to first determine what the requirements are, then to design a system, and implement it. Ideally the system should be on the web so that access and collaboration are easy. ","","Ideal for SE students and those who have dome COMP3111.","",""
"AN09","Enhancing a BURS-Based Code Generating with Register Allocation","AN","Albert Nymeyer","","","Active","BINFBIOMCECSSE","R & D","","BURS is an instruction selector for a compiler. BURS is based on rewriting graphs that represent the program. What BURS does not do is register allocation. In this work the BURS algorithms will be modified to include a mechanism to allocate registers to variables as they are being translated into instructions.","","COMP3131 and COMP4133 are handy.","",""
"AN10","RATS: Research Applicant Tracking System","AN","Albert Nymeyer","","","Active","BINFBIOMCECSSE","R & D","","RATS stands for Research Applicant Tracking System. There are a number of research and development topics possible in RATS, including HCI design issues, hardcore programming, and algorithm design. The system must address human needs and will be used in a real production environment, sosome care will be taken to formulate topics that are useful and well contained. The final system will be a collaboration that involves many academics and students.BACKGROUND: The School of CSE has a set of procedures and a database that together allow it to manage approximately 160 postgrad research students. But the number of students is growing. The school must also offer faster processing of queries, changes and applications, provide better and more informed case handling, be more accountable, and improve the monitoring of student progress so that more students complete Masters/PhD studies, and do so in a shorter time. As well, in the near future UNSW will be subject to a Research Quality Framework (search for it on the UNSW website) that will measure every school's performance in a number of key areas of management. Under these pressures, the current set of procedures cannot cope.MOTIVATION: The most inefficient aspect of the current system is the tracking of requests.Examples of requests are an application for admission, a request to change supervisor, or a request to waiver a course. and applications.After the initial request has been made, there are many places where it can `get stuck', and even be forgotten, as many processes are informal andrely on emails. Someone can simply forget to respond to an email. It is time consuming and error-prone to regularly be retracing correspondence to find out what the current status of a request is. Another major activity of management is regular progress reviews for all students. If a student's progress is not satisfactory, then the student's case may be referred up to the Higher Degree C'tee, which will place conditions on the continuing candidature of the research student. This process involves the school, supervisor and the HDC monitoring the progress of the student over a period of time.Again, it is a tracking problem requiring each of the parties to be involved and respond appropriately. Submitting a thesis is also a management problem: it involves nominating examiners, waiting on examiners' reports,monitoring changes to the thesis and correspondence with the HDC.This process can be made transparent (visible) to all the parties involved.GENERAL DESCRIPTION: RATS will enable research students, academics and the school to immediately know where a request is, and if there is a hold-up, where it has occured. If a student applies for admission, for example, he is asked by the RATS GUI to provide certain (electronic) documents. He submits these documents, and is shown graphically what the `admission process' is in the form of a timeline. A timeline shows the order in which people will handle his request (each represented as a node on a line), and how long (maximally) each node can take. As a request `perculates' along the timeline, it appears as a `flashing token'. The applicant can log on any time to see where his application has reached.The coordinator will have a `to do' list on his RATS GUI. Every day when he logs on, he sees what requests he must approve or reject. He may also elect to `delay' a decision up to the `timeout' prescribed by the system. All these choices are displayed to him as buttons in the RATS GUI.When a decision is made, RATS proceeds with the next step in the process. Every activity/decision is logged of course, and this log can be accessed by people with authority to see the history of any case (this is tracking).The system is like `big brother' in that it controls the progress of all research services provided by CSE.For example, RATS knows when every student is due for their annual review.It sends them a form to fill in, and once they have done so, sends the form onto the supervisor,and thereafter to the chair of the review panel. All this must happen before the date of the review presentation sothat all the paperwork is completed on time.DATABASE: There is of course also a database side to a tracking system: for example, there are student records that containthe names of courses that need to be done, name of supervisor, dates of progress reviews, results of reviews,correspondenece, any scholarships that the student holds, expiry dates and so on.These records are used to generate reports and statistics, and to do forecasting(e.g., RATS should be able to do budget forecasting using the scholarship information).STAKEHOLDERS: There is quite a range of parties (stake-holders) that can potentially be involved in a request.There is of course the student, and the Postgraduate Coordinator and the  Research Admin Officer,but also the HDC, supervisor(s), co-supervisors and other academics, thesis examiners (often overseas),the Graduate Research School and general administration staff.There are also organisations (like NICTA, CSIRO, IBM and CISRA) that have a vested interest in theprogress of research students who are being funded by their scholarships.RATS should allow these organisations access to certain data, but there is a confidentiality aspect here, so security is an issue.CURRENT WORK:The basic idea is to:1) have some graphical representation of CSE's policies and procedures,2) have an `interpreter' convert nodes in the graph into events on a timeline (this is sort of planning)3) have multiple and independent agents each handling specific events on the timelineTogether the agents form a team that implements the processes defined by the CSE pol. & proc. graph.There are a number of RATS topics. Each is a component in the system, or addresses a different aspect. Groups are very welcome.","","While there are many software engineering aspects to the system, there are also algorithm and HCI design questions that mean that both `soft' and `hard' computer science and engineering students could be involved.","",""
"AN12","A WebCMS-Based Software Project Management System","AN","Albert Nymeyer","","","Active","BINFBIOMCECSSE","Development","","WebCMS is a web-based course management system that is used throughout the school. It offers many basic facilities. One facility that it does not offer however is support for project management. A project management tool should allow a group of users to allocate themselves roles, schedule tasks, plan for milestones, communicate with each other, carry out basic QA, have a software repository, and so on. In this work, an existing management tool will be modifed to fit into WebCMS, which itself may also need modification. The integrated system will need thorough evaluation and testing.","","A good knowledge of databases would be an advantage.","",""
"ANF01","Generation of Formal Models for Wireless Network Protocols","ANF","Ansgar Fehnker","","","Active","CSSE","R & D","","Wireless networks and the associated protocols have some interesting properties that make them challenging to analyse them thoroughly.  One important aspect of these protocols is that they are often probabilistic, and that they operate in an environment with random noise and disturbance. Model checking for probabilistic systems is a promising technique that can be employed for analysis of such systems.As the name model checking suggest, the first step for probabilistic model checking is to build a suitable model. In addition, one needs to define a property; this is what we are checking for. Rather than generating the model and the property on a case by case basis, it is the aim to automatically generate those models from a graphical and spatial network descriptions.We have currently a prototype that is capable to generate models for a simple type of protocols, and a simple type of properties. The aim of the project is to extend this class, both in the types of protocols and in the properties that are checked. This all has to be achieved with an eye on scalability. The models should not only be faithful abstractions of the protocols, they should also be such that they can be parsed, checked and evaluated efficiently, even for an increasing size of the network.The aim of the project is develop a library of reusable formal models for wireless network protocols. This library will go accompanied with case studies that illustrate the use of the generated models. The project will lead to a greater understanding of the wireless domain, as well as to familiarity with formal modeling and model checking. ","","","","It is assumed that the student has programing experience, and an affinity with logic. "
"ANF02","Inter-procedural error traces for software verification of C/C++","ANF","Ansgar Fehnker","","","Active","CECSSE","R & D","","This project is situated within the Goanna project  at NICTA. This project develops a tool, also called Goanna, that is able to detect software bugs in C/C++ source code automatically. We started recently to add inter-procedural checks to Goanna. These checks find bugs that occur because of mistakes that are distributed over multiple functions, procedures and files.  This raises the interesting question how to define inter-procedural error traces. These are useful for the user who wants to debug the code, but potential error traces are also used by the Goanna tool itself to improve its analysis. This aim of the project is to develop a suitable definition of inter-procedural error traces, and to integrate them with the different types of analysis, as well as with the user interface. ","","","",""
"ANF03","Inter-procedural interval analysis for software verification of C/C++","ANF","Ansgar Fehnker","","","Active","CECSSE","R & D","","This project is situated within the Goanna project  at NICTA. This project develops a tool, also called Goanna, that is able to detect software bugs in C/C++ source code automatically.  Some of the successful checks of the Goanna tool uses integer constraint solving, to approximate the range of variable. We started recently to add inter-procedural and multi-threading checks to Goanna. These checks looks for bugs that occur because of mistakes that are distributed over multiple functions, procedures and threads. This raises the interesting question how to extend the interval constraint solving techniques to inter-procedural and multi-threading applications.  The challenge of this project is to show that the extension not just work in principle, but also on real source code of several 1000000 LoC. ","","Please contact the supervisor","",""
"ANF04","Inter-procedural error traces for software verification of C/C++","ANF","Ansgar Fehnker","","","Active","CECSSE","R & D","","This project is situated within the Goanna project  at NICTA. This project develops a tool, also called Goanna, that is able to detect software bugs in C/C++ source code automatically. We started recently to add inter-procedural checks to Goanna. These checks find bugs that occur because of mistakes that are distributed over multiple functions, procedures and files.  This raises the interesting question how to define inter-procedural error traces. These are useful for the user who wants to debug the code, but potential error traces are also used by the Goanna tool itself to improve its analysis. This aim of the project is to develop a suitable definition of inter-procedural error traces, and to integrate them with the different types of analysis, as well as with the user interface. ","","Please contact the supervisor for details","",""
"ANF05","Inter-procedural error traces for software verification of C/C++","ANF","Ansgar Fehnker","","","Active","CECSSE","R & D","","This project is situated within the Goanna project  at NICTA. This project develops a tool, also called Goanna, that is able to detect software bugs in C/C++ source code automatically. We started recently to add inter-procedural checks to Goanna. These checks find bugs that occur because of mistakes that are distributed over multiple functions, procedures and files.  This raises the interesting question how to define inter-procedural error traces. These are useful for the user who wants to debug the code, but potential error traces are also used by the Goanna tool itself to improve its analysis. This aim of the project is to develop a suitable definition of inter-procedural error traces, and to integrate them with the different types of analysis, as well as with the user interface. ","","Please contact the supervisor for details","",""
"AS21"," Recognising Lung features in HRCT","AS","Arcot Sowmya","","","Active","BINFBIOMCECSSE","R & D","","This project is part of a larger medical imaging project. The topic of interest is diseases of the lung, viewed primarily through high resolution computerized tomography (HRCT) images, which is a radiologic assessment technique that has become prominent in the last decade. HRCT scans help to detect many lung disease processes that might escape other tests. There are many variations in the images, depending on the stage of the disease and the range of genetic and environment factors. The project goal is to extract features of interest to radiologists, guided by existing medical knowledge. The techniques to be used include segmenting the image into regions of interest, computing region features and applying machine learning and clustering techniques cooperatively, in order to build recognition systems. You will work with real digital HRCT images from a radiology practice.","","AI course desirable","dedicated project machines, existing software, HRCT images available",""
"AS26","Evaluation of Expert Diagnosis for Lung Diseases","AS","Arcot Sowmya","","","Active","BINFBIOMCECSSE","R & D","","High Resolution Computer Tomography (HRCT) has emerged as an important tool for detection and characterization of lung diseases. A computerized diagnosis system that can accurately quantifyand analyse lung abnormalities in HRCT scans would assist radiologists. However, characterizing the performance of medical image segmentation approaches continues to be  a major challenge. Interactive drawing of the desired segmentation (called labelling) by domain experts has often been the only acceptable approach, and yet suffers from intra-expert and inter-expert variation. An approach to compare a collection of segmentations by experts has been unclear so far.In this project, we would like to estimate the 'ground-truth' segmentation from a group of expert segmentations and quantify a measure of the quality for each expert. We would analyse the process of labelling by different experts and propose a new refined label or outline that has minimum intra-expert and inter-expert variation.","http://splweb.bwh.harvard.edu:8000/~warfield/papers/2002/warfield-miccai-2002-appeared.pdfwww.cse.unsw.edu.au/~mithunp","some familiarity with image processing is desirable","project resources include dedicated machines, access to radiologists and wide collection of lung HRCT images",""
"AS43","Feature Extraction in Proteomic Images ","AS","Arcot Sowmya","","","Active","BINFBIOMCECSSE","R & D","","A study of the benefits/failures of the various techniques proposed for feature extraction from proteomic images will be conducted.  The use of multiple get images for feature detection will be explored.","","","",""
"AS90","Adding interactivity to a projected display","AS","Arcot Sowmya","","","Active","BINFBIOMCECSSE","R & D","","In this project we will attempt to add interactivity to a projected display, in particularly at the iCinema Centre for interactive cinema research (see www.icinema.unsw.edu.au). A range of projects on tracking as well as interaction in immersive environments are possible. Contact supervisor for more details.","","AI course desirable","",""
"AYA01","Requirements Engineering for Service Oriented Architecture","AYA","Aybuke Aurum","","","Active","CECSSE","Research","","The purpose of this study to investigate the Requirements Engineering decisions for service oriented architecture application systems","","The student expected to have sound knowledge in the following topics: requirements engineering, service oriented architecture, decision making process","","Student must contact with Dr. A. Aurum Ext 4418"
"AYA02","Value-based Software Development  in Australian Companies","AYA","Aybuke Aurum","","","Active","CECSSE","Research","","The purpose of this study is to investigate Req Eng process in Australian Companies. This is a research topic and student is expected to do reading in the following areas: value-based decisions, literature from economy and marketing as well as software engineering","","john sheppard","","See Dr. A. Aurum on Ext 54418 or send email to aybuke@unsw.edu.au"
"AYA08","Requirement Engineering Process for Mission-Critical Systems","AYA","Aybuke Aurum","","","Active","CECSSE","Research","","student is required to collect data and apply a qualitative/quantitavie analysis.","","sound understanding RE activities, process models","","See Dr. A. Aurum on Ext 54418 or send email to aybuke@unsw.edu.au"
"AYA17","Requirements Engineering for Web-Based Systems","AYA","Aybuke Aurum","","","Active","CECSSE","Research","","Student is required to investigate requirements engineering activities in relation to web based software. ","","good understanding on RE","","See Dr. A. Aurum, aybuke@unsw.edu.au"
"BB01","Integrating Web service Protocol Discovery Tool into ProM Toolset","BB","Boualem Benatallah","","","Active","All Students","R & D","","Web services, as programmatically accessible components over the Web, have gain a wide adoption in the implementation and integration of application in companies. Currently, one hot research issue is to include as a part of service descriptions not only the declaration of its operations, but also description of its behaviour. This latter description is called service protocols. We have a developed an approach that discovers service protocols from information in service execution logs (that contains information exchanged between services). Due to success of this project, it has been requested to be integrated in the ProM (http://prom.sourceforge.net/) toolset, which is a framework for approaches for behavioral model discovery from logs. We need an enthusiastic and diligent student who is eager to explore this exciting new area, and is interested in GUI development and involvement in a world-class research and development project. You will be working in a friendly team of researchers comprising of a professor and senior researchers. Would you need more information, do not hesitate to contact Prof. Boualem Benatallah (boualem@cse) or Hamid Motahari (hamidm@cse). No prior knowledge of Web services is required but familiarity with Eclipse platform is an advantage.","","","",""
"BB02","A Website for Topical Bibliographies","BB","Boualem Benatallah","","","Active","All Students","Development","","opical bibliographies play an important role in academic life. Theycontain list of references of conference papers, journal papers,books and in general any scholarly work organized by relevant topics.They can be rich by having different indexes (for instance byconference, author name, year, ....), rankings (e.g., journal impactfactors, rank of conferences, etc...), and links to relevantlocations on the Web.  An example of a topical bibliography can befound here: http://se-pubs.dbs.uni-leipzig.de/.This project centers around the design and implementation of aWebsite as a general template for topical bibliographies. We need acreative, motivated and interested student looking to make a bigimpact. The tool will be useful for academics, research students, andany research institution to effectively organize, manage and updateall citation in a research area using a topical bibliography. Thechoice of technology and programming language to implement theWebsite depends on the skills and interests of the interested student. We are planning to make available the project as open sourceand free to use for research and academic use on SOC Group Website(http://www.cse.unsw.edu.au/~soc).As a student, you will be working in a friendly mixed-gender tram ofresearchers comprising of an associate professor and p/g students. Ifyou are interested, please contact A./P. Boualem Benatallah(boualem@cse.unsw.edu.au), or Hamid Motahari(hamidm@cse.unsw.edu.au).","","","",""
"BB04","Web Services Security Analysis and Management Tool. ","BB","Boualem Benatallah","","","Active","All Students","R & D","","Web services are programmatically accessible applications published on the Web. Mainly intended for use between enterprises, research in Web services is fuelled by its promise to increase interoperability between systems. Industry giants such as IBM, Microsoft, Sun, Oracle, and BEA are investing heavily in R&D to enable Web services to realize their potential. Currently, the biggest obstacle to widespread use of Web services between enterprises is the lack of de facto standards for security. In particular, authorization is challenging, as the environments are dynamic and volatile.This project centres around the implementation of a trust negotiation framework for Web services. Trust negotiation is an approach to authorization that does not require the user (the service requester) to be known to the service provider before the interaction takes place. Instead of using identity to specify who may access the service, properties of the requesters are used in the authorization. Requesters prove these properties by using digital certificates.We have an existing prototype that has been presented in an international conference of high reputation. Now, we would like build on this success by extending the software with exciting new ideas.Solid experience in Java and/or C# programming is essential. Also, some knowledge of security and/or Web services is useful. For further information please contact the supervisor directly or Halvard Skosgrud (halvards@cse.unsw.edu.au)","","","",""
"BB22","End users Web Services Composition ","BB","Boualem Benatallah","","","Active","All Students","R & D","","Most knowledge workers in enterprises are familiar with Spreadsheets. Spreadsheet users typically perform complex analyses with data sourced from data warehouses or external systems via the Web. In particular, nowadays there are many data services available on the Web, e.g., for currency conversion, which end users can benefit. Integrating these services into spreadsheets is the key to enable users to enjoy their power. This raises the issue of exporting data from spreadsheet to invoke services and also to import them back from the reply of services into the spreadsheet environment.  In this project, we aim to develop a tool for exporting spreadsheet data to external services, which features a mapping language for expressing transformations. This project is in the context of a research platform called Spreadator with our industry partner, SAP Australia, which has been also published in distinguished conferences. We are looking for motivated, enthusiastic and intelligent students who are interested in research and development projects in a cutting edge research area and in particular in end user programming. You will be working in a friendly team of researchers comprising of a professor, senior researchers and p/g students. If you require more information, please do not hesitate to contact Prof. Boualem Benatallah (boualem@cse) or Woralak Kongdenfha (woralakk@cse). ","","","",""
"BB52","GUI for Adapter Development for Web Services","BB","Boualem Benatallah","","","Active","All Students","R & D","","Web services are programmatically accessible applications over the Web and are emerging as the technology of choice for application development and integration. Due to the wide adoption of Web services in companies, there is now need to automatically identify if two or more Web services can collaborate by looking at their (WSDL) interfaces and protocols, i.e., observable behaviors of Web services. We have developed an approach for matching and adaptation of Web services interface and protocol specification, which has been published in distinguished conferences. This approach has been implemented in IBM WID (WebSphere Integration Developer).  In this project, we aim to design a graphical interface for the approach in the IBM WID environment and to evaluate the usability of the tool for end users. This project requires an interested, intelligent and diligent student to join our friendly team comprising of a professor and senior researchers. Would you need more information, do not hesitate to contact Prof. Boualem Benatallah (boualem@cse) or Hamid Motahari (hamidm@cse). No prior knowledge of Web services is required but familiarity with Eclipse platform is an advantage.","","","",""
"BB61","Process Spaceship AJAX-based GUI: A tool for superimposition and management of Process Views over Heteorogeneous IT systems","BB","Boualem Benatallah","","","Active","All Students","R & D","","Due to ever increasing desire for agility in companies, Web services are used in the implementation and integration of various systems in companies. Web services are programmatically accessible applications over the Web that simplify application development and integration. The wide adoption of Web services has given rise to the need to analyze Web services interactions and to understand the relationships between them in such an environment (we call it a process space). We have developed an approach for discovering such relationships (and their business process) in process spaces. In this project, we aim to build an AJAX-based GUI for this system (called Process Spaceship), so that end users can visually browse such relationships (various process views over the process space) and understand them. This project requires an interested and diligent student to join our friendly team comprising of a professor and senior researchers. Would you need more information, do not hesitate to contact Prof. Boualem Benatallah (boualem@cse) or Hamid Motahari (hamidm@cse). For more information on the project see http://www.cse.unsw.edu.au/~hamidm/projects/processspaceship. No prior knowledge of Web services is required but familiarity with AJAX or Eclipse platform is an advantage.","","","",""
"BB63","Spreadsheet based Mashup Tool","BB","Boualem Benatallah","","","Active","All Students","R & D","","End-users, whose primary job responsibilities may not involve programming, in many cases will do simple programming if it makes their life easier. The fact is that so many users are using spreadsheets and its interactive interface for data manipulation and analysis makes spreadsheets a great foundation for creating Web-based composite applications, also called mashups. However, providing a spreadsheet-based mashup tool raises issues of heterogeneities between external data sourced from multiple systems via the Web, as well as between external data and spreadsheet data itself. In this project, we will develop a spreadsheet-based mashup tool for end-users, which features a repository of data integration widgets as the central idea. The widgets include logic for importing, transforming and integrating data from multiple data sources, and presenting it in spreadsheet. This project is in the context of a research platform called Spreadator with our industry partner, SAP Australia, which has been also published in distinguished conferences. We are looking for motivated, enthusiastic and intelligent students who are interested in research and development projects in a cutting edge research area and in particular in end user programming. You will be working in a friendly team of researchers comprising of a professor, senior researchers and p/g students. If you require more information, please do not hesitate to contact Prof. Boualem Benatallah (boualem@cse) or Woralak Kongdenfha (woralakk@cse). ","","","",""
"BG01","Bioinformatics of the immune system","BG","Bruno Gaeta","AC","Andrew Collins","Active","BINFBIOM","R & D","","A number of projects are available in collaboration with A/Prof Andrew Collins, to study various aspects of antibody development. The focus is on the ongoing development of iHMMune-align, a program for identifying and aligning the germline genes that make up rearranged immunoglobulin sequences, and using this program to further our understanding of the mechanisms of antibody-mediated immunityContact supervisor (bgaeta@cse.unsw.edu.au). This project will be a collaboration with the School of Biotechnology and Biomolecular Sciences","Gaeta et al (2007), Bioinformatics 23:1580","This project is best suited to students with experience with statistics and machine learning concepts such as Markov models, maximum likelihood etc. Knowledge of immunology useful but not essential. Some projects may also have a development focus and require excellent Java programming skills","The current version of iHMMune-align and all the publications related to this work are available at http://www.emi.unsw.edu.au/~ihmmune/",""
"BG02","Integrated visualisation of gene networks and microarray data","BG","Bruno Gaeta","","","Active","BINFBIOM","Development","","The goal of this project is to develop an integrated, web-based viewer that allows biologists to visualise the relationships predicted from microarray data together with other attributes of the genes, to evaluate predicted gene relationships and explore the microarray data. Several students have now worked on this application called MINER (Microarray Interactive Network Exploration and Representation) that integrates web technologies, microarray data mining and machine learning, integration of multiple biological annotation sources, and human-computer interaction principles.Contact supervisor (bgaeta@cse.unsw.edu.au)","The background to this project is given in the following publication: Learning Quantitative Gene Interactions from Microarray Data M Bain, B Gaeta, Proceedings of 2nd Australasian Data Mining Workshop, Eds. S. Simoff, G. Williams, M. HeglandUJniversity Technology Sydney, Sydney, 2003, pp. 35 - 49","","",""
"BG08","RNA trans-splicing: artefacts or new splicing mechanisms?","BG","Bruno Gaeta","NSH","Nick Shackel","Active","BINF","Research","","Trans-splicing is a phenomenon involving the formation of novel mRNA sequences through the apparent splicing of exons from different pre-mRNAs instead of removal of introns from one pre-mRNA. Trans-splicing has been observed in a number of cases including disease conditions but is usually explained as the result of sequencing artefacts. The goal of this project is to use bioinformatics methods to investigate the extent of trans-spliced sequences in EST databases and to study whether these are artefacts, ""noise"" from the splicing machinery, or a new biological mechanism to generate sequence diversity. Since trans-splicing is often associated with disease conditions, the project has medical applications.This project is offered in collaboration with Dr Nick Shackel at Royal Prince Alfred Hospital, and includes the possibility of lab work for interested students.","","","",""
"BG09","Phylogenies from 3D structures","BG","Bruno Gaeta","","","Active","BINF","Research","","There is a considerable body of work on deriving phylogenetic trees from DNA and protein sequences, but considerably less on using 3D structure information (with and without sequence) to build phylogenies. The goal of this project is to investigate, develop and evaluate approaches for deriving phylogenetic trees from 3D protein atomic coordinates.  Since structure is more conserved than sequence, structural information should offer a way to build deeper phylogenetic trees.This topic is offered in collaboration with Merridee Wouters at the Victor Chang Cardiac Research Institute","","","",""
"BG10","Escape from the immune system: viral genomes and HLA types","BG","Bruno Gaeta","","","Active","BINF","Research","","This project examines how viral genomes in different populations and MHC alleles in these populations correlate, and how MHC types in a population affect the spread of viruses","","","",""
"BG11","Software for design of PCR primers to detect variation","BG","Bruno Gaeta","","","Active","BINF","R & D","","The goal of this project is to design software to assist biologists in designing PCR primers to amplify variable regions between two closely related genomes. The first part of the project would involve drawing requirements from end-users, followed by design of a suitable algorithm and implementation.","","","",""
"BH04","Real-time Road Traffic Micro-simulation Model and State Estimation ","BH","Bernhard Hengst","","","Active","CECS","R & D","","Smart Transport and Roads (STaR) is a large scale NICTA strategic project. This research topic involves the development of a multi-vehicle micro-simulation traffic model to be used for sensor data-fusion and traffic state estimation. The project will require good programming skills, the ability to implement simulations and animations, and some knowledge of AI. The student will work closely with researchers located on the 3rd floor of the NICTA's Kensington laboratory (UNSW)","","distinction average or above ","","The student will be required to sign a NICTA IP and non-disclosure agreement.  "
"CMK01","Engineering Systems Training Program","CMK","Cat Kutay","","","Active","All Students","Development","","Renewable Energy Systems are coming into more common usage. However the general understanding of these systems is still low. The development and continuous improvement of a comprehensive component modelling system is part of present efforts to improve consumer understanding and predict the behaviour of the various designs for renewable energy systems. Renewable energy systems introduce a high user maintenance and awareness component to energy provision, hence the provision of user learning course and model systems continues to be a high priority of renewable energy system designers and manufacturers. Existing systems are highly technical and do not provide flexibility foradding new components according to a specification file. This project is part of the development of a prototype system which enables users to select thecomponents of a proposed, or existing, energy supply system. The components can be connected into the model by the user and a simulation of thesystem can be run for an extended period.The features of each components are stored in XML files including textual description, attributesa nd image. Users can view various parameter (how much power produced, power available, etc) and any equations describing the functioning of the system. Also warnings are given for incorrect connections, etc.The project looks at the Learning Design. The XML format for the descriptive documents needs to be designed in XML format to distinguish information for different levels of learning (user/maintainer/systemdesigner), also the formulas used in the calcuation of parameters are to be expanded for different levels.","http://www.isi.edu/isd/KRSharing/vision/AIMag.htmlhttp://www.dlese.org/libdev/GUI/taskCDes_InstDesign.html","","",""
"CMK02","Language Learning for Indigenous Languages","CMK","Cat Kutay","","","Active","All Students","Development","","This work carries on developing an existing basic system based around teaching Australian Aboriginal languages. A database structure has been developed and filled, and the Learning interface is in use in NSW schools. A parser has been included to enable students to enter sentences in Wiradjuri and obtain a translation. This will require some parsing and needs to allow for sufffixes and elisions (shortenings) This is being extended for audio recognition. The language has a construction based on limited set of syllables and this could be used to design a very simple language recognitions system.Features we will look at for word development are syllable recognition techniques and elision (shortening) ","Existing dictionary databases available from Cat KutayAlso reference in teaching language by Stan GrantDixon Australian Aboriginal Languages 2002 available in libraray","","",""
"CMK03","Individual Based Model of Group Interactions","CMK","Cat Kutay","","","Active","All Students","R & D","","Computer simulations of groups composed of various individual and interactional features have been used to model biological and environmental systems. These models have also been used in economic modelling. These economic model vary greatly depending on the assumptions of the human caracteristics.This project looks at modelling human individuals interacting in groups or teams, using existing analysis of small group interactions. The focus will be groups working on software development, and the models will be compared with data from existing logs of development groups. The work will involve building the model and adjusting parameters to compare outcomes and then used to predict the cause of different human group interaction outcomes by assuming different conditions.The interaction characteristics will be varied between individual to increase the complexity of the generic group models.","The ""Big five"" and team workhttp://www.it.usyd.edu.au/research/tr/tr581.pdfRange of individual-based modelshttp://www.red3d.com/cwr/ibm.htmlGeneric population modellinghttp://www.stolaf.edu/people/mckelvey/envision.dir/modlog.dir/modlog.html","","Models can be developed on GameMakerhttp://www.gamemaker.nl/Examples:",""
"CR02","Re-designing the G-NAF National Address Database","CR","Chris Rizos","","","Active","CECS","R & D","","The G-NAF (see http://www.psma.com.au) is Australia's nationalgeocoded address file. The project aims to import the G-NAF database andre-design the database structure for faster and reliable performance. TwoDBMSs will be used as the test bed for benchmarking: 1) PostgreSQL plusPostGIS, 2) MySQL with the spatial extension. Oracle 10g is optional.Current G-NAF database works fine, however, there is a room for furtherimprovement such as optimal indexing, corner-address search, designingtables for geometric objects, etc. The outcome of this study is to conformto such a high standard and to be used for a later model of the nationaladdress file.","For more information visit: http://www.psma.com.au References: See student projecthttp://www.gmat.unsw.edu.au/snap/work/2005/cr44-05proj.pdf","Computing/programming/database courses, with skills asappropriate.","DBMSs, and data and database licence will be provided. ","This project has a strong database component, and the studentwill gain valuable experience on an ""industry-strength"" database engine."
"CR03","Software GPS Receiver Development","CR","Chris Rizos","","","Active","CESE","R & D","","The School of Surveying and Spatial Information Systems has developed both asoftware-defined GPS receiver (operating on an Altera FPGA) and a softwareGPS receiver (where all baseband processing is done in software). Severalprojects can be defined for operation on the software platform - creating areal-time receiver design, experiments with different signal acquisitiontechniques, and improved signal tracking algorithms, among others.","OpenSource GPS: http://home.earthlink.net/~cwkelley/SNAP: http://www.gmat.unsw.edu.au/snap/work/theme4.htm","Strong programming ability in C, and interest in RTOS application","","GNSS/GPS software receiver design is currently a ""hot"" topic. This is a challenging project, but ultimately may be very rewarding for the person (or persons) who make real contributions to this development."
"CR11","Using Symbian for GPS-Related Projects","CR","Chris Rizos","","","Active","CECSSE","R & D","","Microsoft's SmartPhone (which is a derivative of PocketPC) and the Symbian OS are emerging as mobile technologies of choice for many product developers. This project is concerned with developing some integrated GPS/mobile telephony applications to run on SmartPhones (incl. 3G type integrated devices).","","","All resources will be provided by supervisor.",""
"CR44","Mobile Applications for the G-NAF National Address Database","CR","Chris Rizos","","","Active","CECSSE","Development","","The G-NAF (see http://www.psma.com.au) is Australia's national geocoded address file. The project will develop the appropriate database structure for storing and accessing this data from a server (where it is resident) to a mobile device wirelessly connected to the server, so that spatial data queries can be made on the G-NAF. Examples of applications that can be developed on a mobile device (e.g. PDA running WinCE OS) include generating maps at locations defined by an address, calculating distances and geometric relations between points defined by addresses, and so on. This is an exciting topic dealing with data that is at the heart of Location Based Services.","See student project http://www.gmat.unsw.edu.au/snap/work/2005/cr44-05proj.pdf","Computing/programming/database courses, with skills as appropriate.","Mobile device, and data and database licence will be provided.","This project offers an opportunity to create a wide range of applications on a mobile device. But it also has a strong database component, and the student will gain valuable experience on an ""industry-strength"" database engine."
"CR49","Location Based Search for User Generated Content","CR","Chris Rizos","","","Active","CECSSE","R & D","","This is a follow-on project to ""Learch"" in which mobile phone users get their coordinates using a GPS receiver, get their point-of-interest by querying a database, and insert their coordinates into the database. Learch is a prototype application that was developed by a 2006 thesis student. This research is intended to enhance the functionality of the prototype, for example, by restructuring and redesigning the database for populating user generated content, implementing Google Maps into the mobile phone application for better visualisation, etc.","This builds on an excellent 2006 project. The thesis & SW can be obtained on demand.","Interest in Location based Services, satellite navigation, mobile devices & database technology.","All resources to be provide by the supervisor.","This project is proposed by the School of Surveying & Spatial Information Systems' Satellite Navigation & Positioning Lab (http://www.gmat.unsw.edu.au/snap)."
"CR60","Image Compression for Near Real-time Web Delivery","CR","Chris Rizos","","","Active","CESE","R & D","","Files containing data collected using high resolution imaging sensors aboard aircraft or satellite are very large in size and, therefore, take a long time to be transported over the internet. For example, the Image Lab in the School of Surveying and Spatial Information Systems has to routinely download such data from the European, Japanese, Canadian and US space agencies, with typical file size of 500MB - 1GB per image. The download usually takes 2-3 hours and has become a bottleneck in delivering near real-time applications with these imagery. This research will study both generic and image-specific data compression techniques in order to dramatically reduce the time needed to transport imagery over the internet.","","Strong programming ability in C++","",""
"CR61","Database Management System for Satellite Imagery","CR","Chris Rizos","","","Active","CECSSE","Development","","Each year the Image Lab in the School of Surveying and Spatial Information Systems obtains hundreds of images from a dozen satellites run by space agencies around the world. In order to support several on-going research projects the effective management of  the image library has become a research topic itself. This project will study the data structures of imagery from different space agencies so that an efficient database management system can be developed tailored for satellite imagery. The system developed should be easy to implement, be scalable and have the potential to be adopted by other research labs and  companies specialising in satellite imaging.  ","","Strong programming ability in C++ and database knowledge","",""
"CR62","Photo-realistic 3D Modelling for Virtual City","CR","Chris Rizos","","","Active","CECSSE","R & D","","In order to prepare better for extreme natural events such as flooding and storm surges, councils of coastal cities are actively building digital models of the built environment - effectively a virtual city - to aid environmental and disaster impact assessment. CAD models are used as building blocks in this process, which often fail to represent the physical city in an easy-to-understand manner because of absence of texture. This project will study the latest techniques for extracting photo-realistic textures from ground-based and airborne photos for 3D city models.","","Suit students interested in computer graphics and visualisation.","","The student has the opportunity to work with officers from the Wollongong City council."
"CR63","High Precision Image Matching","CR","Chris Rizos","","","Active","CECSSE","R & D","","There are many applications in which two images, taken of the same object at different times, have to be matched very precisely, e.g. down to 1/10th of a pixel, so that change between the two image acquisitions can be detected. Current image matching algorithms are generally low in computing efficiency and, in particular, have not taken advantage of dual- or multiple CPUs. This study will seek to develop innovative algorithms for image matching in order to take full advantage of new generation computers.","","","","The student has the opportunity to interact with a group of active researchers and postgraduate students."
"CR64","Database Design for Next Generation GNSS Reference Networks","CR","Chris Rizos","","","Active","CECSSE","Development","","Database schema have been designed for current generation GPS satellites, that store raw tracking data in one RDBMS, collected from a network of continuously operating GPS reference receivers. In the coming years two three trends will become obvious: (1) the number of receivers operating in anetwork(s) will increase to hundreds, (2) the number of navigation satellites will treble, (3) the data types will treble (or more), and (4) the rate of data (& hence volume) will increase (currently data is collected at 1Hz, in future it may be 10x). New RDBMS designs will be necessary. This project will require the design and testing of new database schema, software and operational implementations.","","","","Would suit a student with an interest in database technology."
"ctc02","Environmental data sensing and sharing via mobile phones","ctc","Chun Tung Chou","","","Active","CECSSE","R & D","","People are increasing concerned about their environment. However, government agencies can only provide coarse grain environment data on air quality by setting up monitoring stations at a limited number of places. In order to overcome this limitation, we propose to leverage the ubiquity of mobile phones to collect and share environmental information. In this project, you will build a sensor board which include air quality sensors and UV sensors. This sensor board will communicate its readings via bluetooth to a mobile phone. The phone will then send the information to a portal for sharing. This project requires both hardware and software knowledge. ","Suman Nath, Jie Liu, and Feng Zhao, ""Challenges in Building a Portal for Sensors World-Wide,"" First Workshop on World-Sensor-Web: Mobile Device Centric Sensory Networks and Applications (WSW'2006), Boulder CO, Oct 31, 2006.","","",""
"DRJ01","Technical development process in the XML domain","DRJ","Ross Jeffery","PB","Paul Bannerman","Active","All Students","R & D","","A technical development process is a development process for a particular technology, such as XML, service orientation, object orientation or programming languages.  Such technical development processes are composed of technical steps, best practices, and checklists for different types of technology-specific components at different stages; but these are not necessary aligned with the normal phases of the software development process (such as specifying requirements, design, implementation, testing, maintenance, with some iterations among them). Companies often consider technical development processes as an important addition to their macro-process, bringing a competitive advantage. Some technologies, such as programming languages and object orientation, have not been considered to have a major impact on the normal flow of a development process, but have rather been used for their effect on the efficiency of processes and quality of products.  We observe that some new technologies, such as XML and service orientation, do have major effects on the flow of the development processes. This is largely because such technologies are not confined to the design and implementation phases of a development process. XML has been used to directly define business level requirements and communication standards. Service governance has become part of business activities, since it gives direct control over service development beyond the phases of normal development.   ","contact rossj@cse.unsw.edu.au","","",""
"DRJ13","Service Oriented Architecture for e-Business Standards","DRJ","Ross Jeffery","","","Active","All Students","R & D","","Recently, industry consortiums have been developing e-Business standards using XML and business process modeling. Such standards inevitably have to be mapped on to technology layers such as service-oriented and event-driven infrastructures. Deriving a flexible reference architecture and implementation from e-Business standards are not always straightforward. It involves multiple technical and not-technical factors and design trade-offs. This project will guide students to look into the state-of-art in e-Business standards, web service protocol stacks and service coordination mechanisms. Students will work with one of Australia's leading e-Business standardization body to solve real-world problems by inventing methods and implementing prototypes. The work will contribute to the standardization body directly. Students may also have the opportunity to do additional work, which will be qualified for ""Industry Training"" required by the school. Students will work closely with researchers at National ICT Australia in a very friendly team environment. Suitable for students interested in software design, web services and industry-scale development. For more information, please email: Liming Zhu limingz@cse.unsw.edu.au. This topic will be also co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html).","http://www.research.ibm.com/journal/sj/452/roth.html http://www.research.ibm.com/journal/sj/452/hinkelman.html","  	strong programming skills required.","",""
"DTW10","Software Tools for Chemistry: Molecular Graphics on the Web","DTW","Daniel Woo","","","Active","BINFBIOMCSSE","R & D","","The project involves the user interface design and implementation of software tools that are used by Chemistry. There have been several successful thesis topics that have been carried out in collaboration with the School of Chemistry (and usability tested with Chemistry students). In this series of projects we are looking at an educational application in molecular graphics, providing a web-based learning tool to help educate chemistry students. This year we will be using eye tracking to evaluate the effectiveness of such interfaces.","","COMP3511 Human Computer InteractionWeb Programming","",""
"DTW23","Location-based Application","DTW","Daniel Woo","","","Active","CECSSE","R & D","","This project explores technologies that reveal present location of the user and incorporates this information in meaningful and novel ways. We will be looking this year at a location database of services across the University and a mobile device that can provide information about services that are nearby to the user. the challenges will be to dynamically provide up-to-date information to the users as they roam the campus. ","","COMP3311 Database SystemsCOMP3511 Human Computer Interaction","","PHP, CSS, mySQL, PostgreSQL, .NET on mobile devices"
"DTW35","Developing exergaming systems for falls prevention in older adults.","DTW","Daniel Woo","","","Active","BIOMCE","R & D","","Worldwide, one-third to one-half of the population aged 65 and over experience falls each year, and nearly half of those that fall do so repeatedly.  Nearly 15% of all admissions of older individuals to emergency medical departments are fall related and falls in elderly women occur twice as frequently as do those in men. While not all falls in older people result in injury, even those that are not injurious can have serious consequences for the older person.  Psychological trauma and the fear of falling can have a high impact on the quality of life of the individual that experiences a fall, with self-imposed reduction in activity, mobility and social interaction often resulting.  Most significantly, in Australia, as in the rest of the world, falls are a prominent cause of injury related deaths in people aged 60 and older.Postural instability and falls occur when normal spatial orienting and balance control mechanisms engaged during self-motion through the environment are challenged and fail.   Stepping underpins all human movement through the environment and is the most effective means of maintaining balance, particularly for older people and those with balance and movement disorders.  There is some evidence to suggest that training stepping responses may be an effective means to improve stepping responses in older adults.  As with any repetitive exercise however, adherence to exercise routines remains a significant barrier to effective stepping training in older adults.Video games have recently been increasingly used in health-related fields such as treatment of obesity in children or educating young adults wit cancer about beneficial health activities. Our project aims to develop dance mat style games (StepMania) that will be appropriate for training stepping responses in older adults.  We aim to conduct a clinical validation of this technology in 2009-2010.This project is carried out in conjunction with the Prince of Wales Medical Research Institute.","","Interests in: biomedical engineering, health, exercise","","Assumed knowledge:   Experience in writing computer games, hardware-software interface, data acquisition, networked gaming"
"EM15","Automatic creation of ontology and application of search  techniques to improve interaction with a web-based service provider ","EM","Eric Martin","","","Active","CECS","R & D","","Freelance-market.com is a web-based service provider that allows freelance professionals (eg. programmers, graphic designers, engineers) to describe their profile so that customers can choose the best fitting freelancer for their project. The service is hosted in different countries, including Australia http://www.freelance-market.com, with Germany as the main hub http://www.freelance-market.de/A key feature of Freelance-Market is the search feature that it lets users identify / access the profiles of professionals who could provide the services they are after. At this stage, the search is rather primitive, and there is plenty of scope for improvement. The aim of the project is to develop a search engine that would analyse the data, automatically build an ontology of the domain, and recognize semantic relationships, so that users could have a successful search experience even if the keywords they enter are not used in the professionals' profile, but are semantically related to those descriptions. The system will offer two kinds of search: totally directed, with the user having to select an item from a given menu, or partially directed, the user receiving suggestions to guide an open search whenever the keywords they enter do not improve the current state of the search results. The system will provide users with feedback on how the search progresses, and display results when their number is neither too small nor too large, helping users to either expand or refine the search before that stage is reached. The ontology, semantic relations and search options will be automatically modified as new profiles enter the database and existing profiles are removed from it. The fact that the range of words used in the profile descriptions is not very large and offers many similarities suggests that sophisticated and dedicated search strategies can be designed and implemented that will exploit this particular domain to its fullest.The project will be co-supervised by Eric Martin and Rainer Kurz, who is the creator of Freelance-market.com and director of the company. Data from the web sites and access to the website coding (test web server) will be made available for testing purposes. At this stage, the Australian web-site has a much more limited set of data than the German web site, but the situation will be different when Thesis B starts. Thesis A will be devoted to the design and implementation of search algorithms and ontolgy construction.The project can be either individual or a group project for two students.","","COMP3411","",""
"GH109","Bluesat OS","GH","Gernot Heiser","","","Active","All Students","R & D","","Design and implement an operating system for Bluesat","","","",""
"GH110","Multi-kernel fault tolerance","GH","Gernot Heiser","","","Active","All Students","Research","","Fault tolerance of computer systems is important in mission-critical applications. A number of  standard approaches to fault tolerance exist,  including hardware- and software redundancy.  However, protection from redundancy-based  aproaches is limited, as long as the redundant  components are identical, and thus likely to fail  in exactly the same way at the same time. A more  reliable approach to redundancy-based fault  tolerance would employ components that are  independently engineered to the same (or  compatible) specification.While this is a reasonable approach at the application software level, the cost of independently manufactured operating systems is normally prohibitive. Given that failure of the operating system is generally fatal for the system, this implies serious limitations for practical fault tolerance.However, the small size and simplicity of a microkernel could make kernel-level redundancy feasible. Furthermore, on some architecturesthere already exist independently engineered implementations of the L4 microkernel, namely L4Ka::Pistachio and Fiasco. This thesis is toinvestigate the use of these independent kernels as a redundant fault-tolerant operating systems platform on a suitable SMP or SMT system, and design and implement a prototype.If successful, this work can lead to a landmark publication.","http://l4hq.org/http://www.disy.cse.unsw.edu.au/","Strong OS background. Distributed systems background is desirable","KEG",""
"GH115","Optimiser for Itanium System Code","GH","Gernot Heiser","","","Active","All Students","Research","","Low-level Itanium operating system code, such as TLB miss handlers, often needs to be written in assembly code. Instruction scheduling is explicit in Itanium assembly code, so good scheduling is time-consuming, makes the code difficult to read, and requires a separate implementation for each processor model. We suggest a better solution would be to write the assembly code in ""unbundled"" form, and implement an optimisation tool which uses some appropriate constraints and heuristics to decide how to order and bundle instructions.","","Good understanding of machine model and assembly programming","KEG",""
"GH116","Extend SKI/vNUMA to support unmodified kernels","GH","Gernot Heiser","","","Active","All Students","R & D","","HP's SKI Itanium simulator, as well as our vNUMA virtual machine, currently use a simplified device and configuration interface, and require a separate Linux kernel to one compiled for real Itanium hardware. This project would involve extending SKI and/or vNUMA to simulate a few basic devices and hardware configuration tables, so that an unmodified operating system kernel can be supported. This might also be useful to support kernels where the source is not available (such as HP-UX and Windows).","","Good OS background","KEG",""
"GH118","Sunswift components","GH","Gernot Heiser","","","Active","All Students","Development","","A number of Sunswift issues that can be tackled by a thesis student include:    * Improved CAN nodes (less power consumption, more isolation, utilise improved controllers to improve connection reliability)    * Design and build tyre-pressure sensors    * Reliable wireless communication with support vehicleThe actual thesis topic would be decided depending on the skills and interests of the student. ","","Interest in hardware, ability to use a soldering iron and multimeter (among others...)","KEG",""
"GH122","T-Kernel on L4","GH","Gernot Heiser","","","Active","All Students","R & D","","μITRON is an API specification for a real-time OS kernel that is widely used in the Japanese and Korean industry. The basic μITRON specification is aimed at MMU-less micro-controllers. An implementation of this ``unprotected'' μITRON on L4 has recently been done by a local student.In practical terms, however, running L4 on hardware without memory protection is pointless, and hence an implementation of unprotected μITRON on L4 is of limited practical use. There is a ``protected'' version of the μITRON API (for microprocessors with memory protection), but it is presently available in Japanese only. Moreover, it seems that it has found little uptake in industry. Instead, a new standard has emerged in the TRON family: the T-Engine, which is a standardised embedded platform, with a standard OS, the T-Kernel.This project will analyse the T-Kernel specification, understand how it can be mapped to L4, and design and implement an L4-based T-Kernel prototype. This may be based on the existing μITRON implementation.","http://www.ertl.jp/ITRON/SPEC/mitron4-e.htmlhttp://www.t-engine.org/english/whatis.html","Excellent OS background","KEG",""
"GH124","Elan compiler for L4/Iguana","GH","Gernot Heiser","","","Active","All Students","Development","","Elan is an elegant and high-level language which was extensively used in L3, the predecessor of L4. Source for an Elan compiler for L3 (written in Elan) is available. Availability of an Elan compiler is a pre-requisite for migrating existing commercial L3 installation to the L4/Iguana operating system.This project is to port the Elan compiler to Linux and re-target it for L4/Iguana. As the compiler itself is written in Elan, this will require bootstrapping it on the new environment.","http://os.inf.tu-dresden.de/L4/l3elan.html","Good programming ability, interest in compilers","KEG",""
"GH125","Fault-tolerant embedded system","GH","Gernot Heiser","","","Active","All Students","R & D","","Microkernels provide hardware-enforced encapsuation of system components. This provides fault containment and a basis for fault tolerance.This thesis is to inviestigate approaches to component-level fault tolerance applicable to the L4/Iguana embedded operating system and the CAmkES component architecture. It is then to design, implement, test and benchmark a fault-tolerant prototype demonstrating the viability of one or more of those approaches.","","Good understanding of operating systems concepts, good programming ability","KEG",""
"GH128","Performance limits of Darwin on L4","GH","Gernot Heiser","","","Active","All Students","R & D","","Darwin is the kernel of the Mac OS X operating system. It consists of a Mach kernel with a co-located BSD system. Darwin has recently been ported by the UNSW/NICTA OS group to run as an unprivileged (para-virtualised) guest on the L4 microkernel.The present L4/Darwin port (Darbat) is a proof-of-concept prototype. This thesis is to analyse Darbat's performance, identify bottlenecks, and resolve them by re-design/re-implementation, in particular by reducing the dependence on the poorly-performing Mach code. The aim is to establish the inherent performance limits of the approach compared to native Darwin, and in particular where it is inherently worse and where performance is improved. Attempt to draw conclusions about the performance of an L4-based Mac OS X.","","","",""
"GH129","Afterburning Linux on Itanium","GH","Gernot Heiser","","","Active","All Students","R & D","","Afterburning, also called pre-virtualisation, is a way to semi-automatically virtualise an operating system to run in a virtual machine. Unlike the better-known para-virtualisation technique, the result is an OS binary that can run on any hypervisor as well as on bare hardware. Yet, pre-virtualisation achieves the same performance as para-virtualisation.Pre-virtualisation has been fully implemented on x86 hardware in a collaborative project between NICTA, UNSW and the University of Karlsruhe. On Itanium, the NICTA/UNSW so far has produced a partial implementation, that does not yet produce a binary that can run on hardware as well as any hypervisor. There are a number of reasons why the x86 approach doesn't easily translate to Itanium.This thesis is to design, implement, benchmark and optimise full pre-virtualisation on Itanium, for bare hardware and the Xen and vNUMA hypervisors. If time allows, the L4 microkernel and Linux can also be supported as hypervisors.","http://ertos.nicta.com.au/research/virtualisation/http://l4ka.org/projects/virtualization/","Excellent C programming skills, good OS background, ability to deal with large and complex systems, interest in computer architecture","KEG",""
"GH132","Crashable I/O Kit Drivers","GH","Gernot Heiser","","","Active","All Students","R & D","","Darbat is a version of Darwin, the kernel of Apple's Mac OS X operating system, on the L4 microkernel. I/O Kit is the device driver framework for OS X, including the actual drivers. Darbat presently runs all I/O Kit device drivers in a single user-mode process on L4.This thesis aims to run multiple I/O kit processes on a single machine, in order to confine device drivers. Part of this thesis will involve measurement of different system architectures to evaluate the trade-offs in user-level I/O kit device drivers. The major goal is to allow a system to continue with minimal disruption when a non-essential device driver crashes. ","http://ertos.nicta.com.au/software/darbat/","Good OS background, excellent C programming skills","KEG","cgray"
"GH133","Darbat Aqua","GH","Gernot Heiser","","","Active","All Students","R & D","","Darbat is a version of Darwin, the kernel of Apple's Mac OS X operating system, on the L4 microkernel. Currently Darbat supports a single-user shell prompt as well as user input device drivers, however it can not run the full OS X GUI.This project involves putting in the necessary features to the Darbat virtualisation layer to support running OS X GUI applications. This will require working at the L4 level as well as Darwin kernel hacking. ","http://ertos.nicta.com.au/software/darbat/","Good OS background. Excellent C programming skills","KEG","cgray"
"GH134","Darbat Resource Management","GH","Gernot Heiser","","","Active","All Students","R & D","","When virtualising an operating system, getting it working is only half the story. Darbat is a para-virtualised version of Darwin, the kernel of Apple's Mac OS X operating system. Darbat runs on the L4 microkernel. Several Darwin instances can run concurrently on the same L4 kernel.This thesis will involve examining resource management issues in Darbat. It will require examining static policies such as partitioning as well as dynamic techniques for scheduling and VM management. An analysis of the trade-offs involved between performance and security will be required. Optionally, this project may involve building a user interface for runtime management of resources. ","http://ertos.nicta.com.au/software/darbat/","Good OS background. Excellent C programming skills","KEG","cgray"
"GH135","Forensic-friendly operating system","GH","Gernot Heiser","","","Active","All Students","Research","","Computer forensics deals with extracting information from computers after crimes (be it that the computers were used as a tool by a criminal, or that the computer itself is the crime scene). It also is about the design of computer systems to support forensic analysis.This thesis investigates microkernel-based system designs in this context. For example, given formally verified microkernel (that has been shown to be secure against attempts to compromise it), how can this be leveraged into constructing a system where it is impossible to destroy evidence of break-ins.The thesis will survey the relevant literature of computer forensics, propose a design of a forensic-friendly microkernel-based system, and implement a prototype.","http://www.cit.uws.edu.au/compsci/computerforensics/","Very good understanding of OS and security issues","KEG",""
"GH137","Maximum power point tracking for the Sunswift 4 solar race car","GH","Gernot Heiser","","","Active","All Students","R & D","","Maximum power point trackers allow a solar panel to operate under its optimal operating conditions. They are a power electronics device which converts between the panel's optimal operating voltage and the output voltage which is determined by the battery state of charge. MPPTs have generally been a weak point of the Sunswift race cars. The team have designed a highly efficient (>98%) MPPT, with reliable hardware, but the control software has a number of issues which causes the devices to behave sub-optimally.This project would, depending on the student's aptitude, first of all perfect the control software for the MPPTs in Sunswift 3. Secondly, implement new features such as diagnostics (IV curve sweeps) and improved tracking algorithms (aiming to improve the solar array power output), and explore efficiency-increasing behaviours (e.g. synchronous rectification). Lastly, depending on the available time, the student would look at implementing improved MPPT hardware based on a novel power-electronics architecture.This project would require close collaboration with the UNSW solar racing team","http://www.sunswift.com/","","",""
"GH138","On-race strategy and data analysis software for the Sunswift 4  solar race car","GH","Gernot Heiser","","","Active","All Students","R & D","","Modern solar racing cars have a limited battery storage (equivalent to about a third of a day's energy needs, running at 100km/h). This charge must be managed over the course of a race in order to complete the race most efficiently (similarly to the way a a Formula 1 team must manage the car's fuel through the duration of a race). Additionally, real-time telemetry data from the car must be analysed to determine the car's correct operation, and the severity of any faults. Data about the course, data being transmitted from the car itself, data from sensors around the support fleet (wind speed/ direction, solar radiation, road conditions, etc), and anecdotal data from the support crew can all be taken into account.This project would first of all develop models for the behaviour of a solar car. A testing scheme would be developed and data gathered (via on-road testing) to characterise the models. A software system would be developed to analyse data coming from the car with respect to these models. Finally, methods for choosing the speed at each point along a pre-surveyed course would be implemented. The entire package should be tested on-road with Sunswift 3 in preparation for future solar car races. Depending on the success of the strategy decisions, the optimal speed could be fed back to the car's cruise control system in real-time, allowing for intelligent, efficient, autonomous cruise control.This project would require close collaboration with the UNSW solar racing team.","http://www.sunswift.com","","",""
"GH139","Accurate data acquisition for the Sunswift 4 solar race car","GH","Gernot Heiser","","","Active","All Students","R & D","","The Sunswift solar car has the most advanced electrical system of any solar racing car in the world, with fully-custom electronics feeding real-time telemetry information to the support crew during the race. The car has a 30kg Lithium Polymer battery pack running at up to 164V. Estimating the charge left in the batteries is crucial to implementing good strategy during the race (i.e. never over-filling, or running the batteries flat except at the start and end of the race respectively).This project would focus on implementing a battery-state-of-charge estimation system for Sunswift electrical system. The system would connect to the car's control area network (CAN), allowing these measurements to be read by the driver and wirelessly by the support crew. While the exact approach would be dependent on the student, one method requires an integration of the current into and out of the battery, along with some linear modelling. Ideally, other factors, such as the voltage, would be determined to provide multiple methods of characterisation. Depending on the aptitude of the student, a large number of other useful sensors could and should be developed, including tyre pressure, temperatures, tilt, acceleration, etc. all connected to the car's telemetry network, feeding information to the driver and support crew.This project would require close collaboration with the UNSW solar racing team","","","",""
"GH140","Steering-wheel integrated driver controls and display for the Sunswift 4 solar race car NEW","GH","Gernot Heiser","","","Active","All Students","R & D","","Recent changes to the international technical regulations for solar racing were required to slow the cars below the road speed limit. One new regulation requires a steering wheel. By integrating the driver controls and display into a steering wheel, all driver interfaces with the car can be consolidated in a similar way to Formula 1 racing cars. In addition, the greater visibility and area for controls allows for more functionality to be built into the steering wheel, including all driver controls, data logging, speed control, and a graphical display.This project will involve the specification, design, manufacture, integration and testing of steering-wheel based computer for a next-generation solar racing car. This will require close collaboration with the UNSW solar racing team. An ideal student would have, or have the ability to very quickly learn, PCB design and low-level software skills. ","http://www.sunswift.com/http://www.wsc.org.au/","","",""
"GH141","Analysis of Linux kernel bugs","GH","Gernot Heiser","","","Active","CECSSE","Research","","Building new, more reliable, operating systems requires good understanding of the sources of failures in current systems.  To date, very little systematic knowledge has been collected about types and causes of bugs in operating systems. This project will give answers to the following questions: What types of bugs are most common in system code?  How and why are particular types of bugs introduced?  What is the typical lifecycle of a bug?<br>Initially, the study will be confined to a single part of the Linux kernel &mdash the USB device framework.  The project will involve in-depth study and reverse engineering of Linux kernel code.  In addition, development of software tools for automatic code analysis may be required.<br>Results of the project will be incorporated in a research paper that will be submitted to one of the major conferences on operating systems or software reliability.<br>FILLER to defeat the silly word-count rule FILLER FILLER","","","","Strong C programming skills required.OS background advisable, but not essential.Software-engineering background useful"
"GH142","Analysing the L4 IPC fast path","GH","Gernot Heiser","","","Active","CECSSE","R & D","","The high-performance OKL4 microkernel uses an assembler ÒfastpathÓ to overcome the performance problems resulting from C code in the critical IPC system call path. Relying on assembler code has a high engineering and maintenance cost, and makes formal verification harder. Eliminating the need for assembler code, or, at least, reducing the amount needed, would have significant practical benefits.<br>This thesis is to analyse the IPC path in order to understand why C compilers do not do a better job on it. It will analyse possible (semantically-invariant) modifications of the C code, in order to investigate how far C performance can be pushed, and whether it can be made competitive with the assembler implementation. This study is to be done on at least two architectures (x86 and ARM) and using several compilers (gcc, Intel compiler, RVCT, maybe Green Hills).<br>A thorough and insightful evaluation should be publishable in an OS workshop.","","COMP3231","","Good C and assembler programming skills, Basic understanding of computer architecture. Aptitude for low-level programming."
"GH143","DTrace on OKL4","GH","Gernot Heiser","","","Active","CECSSE","Development","","DTrace is a dynamic instrumentation facility originally developed for Solaris. It that allows users to augment operating-system execution with user-defined code. This thesis is to design and implement a Dtrace facility for the OKL4 microkernel.When dormant, DTrace probes should present no performance overhead. When active, they allow the user to analyse performance profiles, call paths, and identify potential bottlenecks in the OKL4 kernel. Given the extremely short duration of most OKL4 system calls, meeting the no-overhead is a challenge.<br>The goal is to provide support for the following DTrace providers (or equivalent) in OKL4: lockstat, fbt, sdt, syscall, vminfo, and sched. Performance and usability of the implementation is to be evaluated. Ultimately you want to demonstrate whether such a facility makes sense for a microkernel.","","COMP3231, COMP9242","","Excellent OS background, excellent C programming skills, experience with assembler."
"GH144","Disk scheduling with latency guarantees","GH","Gernot Heiser","","","Active","CECSSE","Research","","Modern disk controllers re-order disk requests in order to maximise throughput. This means that on a highly loaded disk, individual requests can experience significant latency, as much as 3 seconds. This is unacceptable for many web servers.<br/>This thesis is to evaluate existing approaches to disk scheduling with QoS guarantees, and determine which form the best starting point for a web server scenario. Design a suitable approach, and implement it in Linux. The aim is to get a highly probably (say 99%) worst-case response time of the order of 100ms.<br/>This research is a collaboration with Google.","","COMP3231, preferrably also COMP9242","","Strong OS background, excellent C programming skills, strong analytical skills"
"GH145","Linux multi-processor scalability","GH","Gernot Heiser","","","Active","CECSSE","Research","","Linux is claimed to be scalable to 1000s of CPU. However, experience shows that this is only really true for HPC workloads that do essentially no system calls.<br/>This thesis is to analyse Linux scalability, identify bottlenecks and suggest approaches to resolve them. Specifically develop and run benchmarks that perform concurrent system calls on multiple processors, and measure they overhead as a function of the number of processors. Use profiling to determine the bottlenecks.<br/>This work is in collaboration with HP and is expected to lead to publishable results.","","COMP3231, preferrably also COMP9242","","Strong OS background, excellent C programming skills, strong analytical skills"
"GH146","X server for OKL4","GH","Gernot Heiser","","","Active","CECSSE","R & D","","In recent years the X windowing system, originally designed for desktops, has become more prevalent on embedded system. This is mostly driven by the widespread use of X, provision of an X server therefore significantly increases the amount of software that can be easily ported to an embedded platoform.<br/>The OKL4 microkernel operating system is becoming widely used in medium- to high-end embedded systems, including mobile phones and other consumer electronics, network infrastructure devices and others. While it can host a virtualized Linux system, a lightweight X server that does not depend on an underlying Linux kernel would be an attractive service for devices that do  not have megabytes of memory.<br/>The aim of this thesis is to port an appropriate X server to OKL4. This would involve rewriting the underlying communication mechanism to use OKL4 primitives, rather than socket-based communication. The resulting system is to be evaluated against native Linux.","","COMP3231 or equivalent (essential), COMP9242 (desirable)","","Essential: excellent C programming skills, ability to cope with large and complex software systems, very strong OS backgound."
"GH46","Port L4 to SuperHitachi","GH","Gernot Heiser","","","Active","All Students","R & D","","Port L4KA::Pistachio to the SuperHitachi architecture. This will cover the last outstanding mainstream general-purpose architecture and make L4Ka a truly platform-independent kernel. Serious operating system kernel hacking...","http://l4ka.org","Excellent OS background, COMP9242, excellent low-level programming skills, good understanding of architecture","KEG lab",""
"GH85","L3 on L4","GH","Gernot Heiser","","","Active","All Students","R & D","","# L3 is a persistent operating system which has been in commercial use for 20 years. It is also in a sense the predecessor of L4. However, L4 is a very minimal kernel, which requires general system services, such as those provided by L3 (or UNIX) to be provided by user-level servers.The existing user base of L3 presents a potential user-base for an L4-based system, provided a (more-or-less) L3-compatible system could be provided on top of L4. The purpose of this thesis is to examine the structure of L3, design an L3 emulation on L4, and implement a first prototype. Schšnbeck, the present L3 maintainers, have agreed to help.","http://l4hq.org","Excellent OS background","",""
"GH89","Untrusted Device Drivers","GH","Gernot Heiser","","","Active","All Students","Research",""," The BE thesis of Ben Leslie has recently developed and implemented a device driver model for the Mungi operating system which runs drivers at user level without special privileges. With the help of hardware mechanisms available on some contemporary computer systems, the work could even limit the damage a driver could do via direct memory access (DMA). This means that drivers need not be trusted by the system, unless they perform an I/O operation on whose correct and timely execution the security of the system depends.This thesis is to develop a system model which avoids trusting disk and network drivers. The work is to be performed for Mungi (as a relatively small and well-structured system) but results should, in principle, be applicable to other systems (such as Linux).","http://www.disy.cse.unsw.edu.au/ULDD/","Excellent understanding of OS issues, excellent programming skills.","",""
"GH94","JVM on L4","GH","Gernot Heiser","","","Active","All Students","R & D","","Port a Java virtual machine (JVM) to L4. This will provide a minimal Java environment for L4-based embedded applications and constitutes an important step into making L4 an ideal base for building embedded systems.","http://www.disy.cse.unsw.edu.au/http://l4hq.org/","Good OS background and programming skills","KEG",""
"GH96","Sunswift OS","GH","Gernot Heiser","","","Active","All Students","R & D","","Design and implement an L4-based operating system for Sunswift, the UNSW solar racing car. Sunswift features a CAN-based control system with a PLEB for central coordination and external communication. The PLEB presently runs Linux.This project will build a new, lightweight OS for Sunswift, specially designed to meet the car's requirements for real-time control, communication and reliability. It will be an initial prototype and demonstrator for an L4-based framework for embedded systems software.","http://www.disy.cse.unsw.edu.au/http://l4hq.org/","Suitable for a student with a good OS background, L4 experience and an interest in embedded systems and interfacing to hardware.","KEG",""
"GH99","Multics on modern hardware","GH","Gernot Heiser","","","Active","CECSSE","Research","","Forward-port Multics to Itanium or 64-bit PowerPC. Evaluate performance and compare to Mungi.<br>While this topic had originally been posed as a joke (which not everybody got), the recent <a href=""http://www.kirps.com/web/main/_blog/all/mit-releases-the-sources-of-multics-the-father-of-unix.shtml"">release of Multics sources</a> means that this could be looked at seriously. Nevertheless, it isn't for the faint-hearted!","","","","Some OS prior knowledge required ;-)"
"GWK02","Verifying the core of standard C library in Isabelle/HOL","GWK","Gerwin Klein","","","Active","CSSE","R & D","","You will work with a state-of-the-art interactive theorem prover (Isabelle/HOL) to formally verify the functional behaviour of a small number of basic C functions like memcpy, memset, etc. The verification of these functions is at the basis of any undertaking that wants to provide guarantees about programs implemented in C.This project is an integral and important part of the formal verification of the L4 micro kernel at NICTA. You will work with the developers of L4 and Isabelle in an international team of PhD students and researchers in the NICTA Embedded, Real Time, and Operating Systems program and the NICTA Formal Methods program.","","An good understanding of C and some background in formal logic is essential. Experience with Isabelle/HOL or having heard COMP4161 is not required, but would be beneficial.","","You will work as part of a significant research project with a team of international PhD students and researchers at NICTA Kensington."
"GWK04","Translating Haskell to Isabelle/HOL","GWK","Gerwin Klein","","","Active","CSSE","R & D","","Haskell is a pure, side-effect free functional programming language that is well suited for formal reasoning.The topic of this thesis is translating Haskell programs into higher-order logic to support formal verification of Haskell programs in Isabelle/HOL.Interesting complications in this translation process are that functions in HOL are less expressive than in Haskell (only terminatingfunctions vs lazy evaluation and partial functions), that the type system of HOL is less expressive than the one of Haskell (simple typeclasses vs monads and constructor classes), and many convenient syntactic language concepts of Haskell are not available in Isabelle/HOL (deeply nested patterns, guards in recursive definitions,etc).Nevertheless it is possible to translate large parts of the Haskell98 standard into Isabelle/HOL, including a significant percentage of ""real-life"" Haskell programs. A benchmark for the quality of the translation will be the Haskell prototype of operating system microkernel seL4, developed by National ICT Australia. There currently already exists an automatic translation from this  prototype into Isabelle/HOL achieved by simple text-processing in python. The purposeof this project is to implement this translation properly in Haskell or ML, to generalise it to a larger set of programs, and to (informally) show the correctness of the translation process.","","The ideal would candidate would have a strong background in compiler construction and functional languages.","","You will work in an international group of PhD students and researchers in the Formal Methods program in the Sydney/UNSW Laboratory of National ICT Australia."
"GWK05","A Take-Grant Security Refinement Proof for seL4","GWK","Gerwin Klein","","","Active","CSSE","Research","","The take-grant model is an abstraction of systems that use the concept of capabilities to implement access control. A capability is a reference to a subject of the system, together with a set of rights. If subject A has the ""grant"" right on a capability pointing to B, it can transfer other capabilities to B (""take"" is the dual operation). Other rights include the traditional reading/writing of data.National ICT Australia is currently developing seL4, an operating system microkernel with these characteristics, and aims to verify its correctness in Isabelle/HOL. Establishing the take-grant model for seL4 is part of this effort.The topic of this thesis is to extend an existing adaption of the take-grant model to seL4 in the theorem prover Isabelle/HOL and proof that the existing formal model of seL4 formally refines this take-grant security model.","","A strong background in formal logic is required. Experience with Isabelle/HOL is recommended.","","You will work in an international group of PhD students and researchers in the Formal Methods program in the Sydney/UNSW Laboratory of National ICT Australia."
"GY01","Automatic traffic monitoring using video cameras ","GY","Getian Ye","","","Active","CECSSE","R & D","","There has been a growing interest in automatic traffic monitoring using video cameras because they provide more visual information by comparison with other sensors. This project is to develop video processing techniques for real-time traffic monitoring. It will involve: (1) fault detection; (2) stop and moving vehicle detection; (3) traffic speed estimation; (4) robust vehicle tracking. This is a challenging and exciting project. You can choose one of the above as your thesis topic. You will have an opportunity to access our testbed system and will be given clear guidance. You will work in Multimedia and Visual Communications Group, Kensington Laboratory, National ICT Australia. This project will improve your skills in problem solving and software development.","","C/C++","",""
"GY02","Detecting objects in images","GY","Getian Ye","","","Active","CECSSE","R & D","","Automatic detection of objects is an important image understanding task with many potential applications, for example, car detection for traffic video analysis. The main difficulty in object detection arises from the large variations in appearance of an object. In addition, the same object can appear very different under different lighting and viewing conditions. A robust object detection system should be able to detect the presence or absence of an object under different poses, scales, and illuminations. This project aims to develop a robust approach to object detection using image processing and machine learning techniques. You will work in Multimedia and Visual Communications Group, Kensington Laboratory, National ICT Australia. Clear guidance and supervision will be provided. Your technical and programming skills will be greatly enhanced by doing this project.","","C/C++","",""
"GY04","Recognizing Human Activities without Supervised Learning","GY","Getian Ye","","","Active","CECSSE","R & D","","The recognition of human activities from video streams has several applications, including virtual reality, smart monitoring and surveillance systems, as well as motion analysis in sports. Most existing techniques use supervised learning, i.e., the traning data has to be manually labeled by users beforehand. This project aims to develop a vision-based technique for recognizing human activities in the surveillance domain without using supervised learning. You will work in Multimedia and Visual Communications Group, Kensington Laboratory, National ICT Australia. This project will improve your skills in problem solving and software development.","","C, C++","",""
"HYP02","A Tool for modeling privacy-aware Web service protocol","HYP","Helen Paik","","","Active","All Students","R & D","","One of the problems in addressing privacy issues in Web services is that there is no proper modelling technique for capturing the privacy requirements for a Web service. That is, no current Web service modelling technologies offer a simple way to state a privacy requirement (e.g., The intended recipient of this message is a delivery service and the data should be removed after the delivery is completed.) in a Web service model.<p>Privacy policies do not discuss the behaviour of individual business applications within the organisation that actually collect/analyse and distribute personal data. This makes the enforcement of the policies difficult. We argue that a model- driven approach, where privacy policies are modelled explicitly as part of the Web service behaviour, can contribute to making the privacy policies explicit and enforceable.In this project, we propose a Web service modeling technique and its supporting tool purposely designed to capture privacy abstractions while describing the behaviour of a Web service. The tool will provide the Web service protocol designer a way to annotate a Web service protocol with various privacy properties. The tool should also verify that the annotated model is correct with regards to some pre-defined criteria.","ServiceMosaic Web site (http://servicemosaic.isima.fr)Boualem Benatallah, Fabio Casati, Farouk Toumani, Julien Ponge, Hamid R. Motahari Nezhad: Service Mosaic: A Model-Driven Framework for Web Services Life-Cycle Management. IEEE Internet Computing 10(4): 55-63 (2006)Boualem Benatallah, Fabio Casati, Farouk Toumani, Rachid Hamadi: Conceptual Modeling of Web Service Conversations. CAiSE 2003: 449-467 ","","",""
"HYP11","Towards End-User Composition Platform","HYP","Helen Paik","","","Active","BINFCECSSE","R & D","","With the large amount of information and services available on the web,it is sometimes difficult for regular users to piece it all together.With Web Service technology, it is possible to integrate businessfunctionalities into one portal service, which consumers could greatlybenefit from.However, most portals stop short of providing their customers with theability to aggregate services on-demand. They may support a simple,pre-defined workflow (e.g., book flight, rent a car, then bookaccommodation), or enable personalised configuration of individualservices (e.g., Weather service for Sydney), but regular users cannot'compose' the service functionality on offer as a need arises.Service composition can create new functionality beyond the individualcomponents and help enhance user experience. Although servicecomposition has been a well-discussed topic for the researchers anddevelopers of Web services, the tools and methodologies for enablingend-user service composition have been largely ignored.Mashups enable users to aggregate and filter information from more thanone source at one convenient location. Intuitively, the concepts Mashupsrepresent could be applied, not just to data but also to Web services.If business had their operations exposed through web services, consumerscould utilise these services and engage them into a process to suit theindividual's needs.In this project, inspired by the ideas in mashups, we attempt to providean example implementation of how end users can benefit from Web servicescomposition.  Students are welcome to contact the supervisors to tailorthe details of individual thesis projects within the given context.","","","",""
"HYP12","Towards Personal Process Management","HYP","Helen Paik","","","Active","BINFCECS","R & D","","Despite huge technology advances, the power of software reuse has not          been transferred to the daily activities of the computer end-users Inthe business or personal computing environments, where end users     operate, people carry out repetitive tasks on a daily basis.  We also          see that, in such an environment, the knowledge gained by the peoplecarrying out the routines is locked away (e.g., in a person's email   folder, or simply in someone's  head), inaccessible to others.                                                                     That is, even though we have sophisticated communication means      available, we have no easy way of reusing or sharing our ``knowledge''about how we do what we do.  We believe that this is due to the lack ofsupport for a systematic approach for those end-users to effectively   define, automate, reuse and share the tasks.                                                                                                                 We use the term ``personal processes'' to refer to daily, repetitive            tasks A personal process can be as simple as opening a computer account        for newly arrived personnel, a finding a restaurant to go out for aFriday night and make a reservation, or as complicated as preparing an        annual tax return.  Common characteristics of personal processes are    that they consist of a sequence of ``instructions'', which allows them  to be described and executed.                                                                                                             Under this project, we will develop a system                                    which is a new and innovative platform designed for personal process management.  This is an umbrella project which consists of manysmall projects. Students are welcome to contact the supervisorsand discuss the scope and shape of individual thesis project for themselves.","","","",""
"IK10","Click Modular Router on L4","IK","Ihor Kuz","","","Active","CECS","R & D","","Investigate, design, and implement a <href=""http://www.pdos.lcs.mit.edu/click/"">Click</a> compatible modular network router architecture on L4 making use of the <a href=""http://www.nicta.com.au/research/projects/component_architectures_for_microkernel-based_embedded_systems"">CAmkES</a> component framework.Click is a software architecture for building network router software from small, reusable, software components, while CAmkES is acomponent-based framework for developing L4-based systems.  Given the componentised nature of both, it should be possible to define a Click router in terms of CAmkES components.  The project will require you to design a framework on L4 that will allow Click components to be reused to build network routers.  Besides designing and implementing this framework, you will also reuse existing Click components to build several variations of network routers and compare the performance of your implementation to existing Click implementations.","","","","See http://www.ertos.nicta.com.au/education/theses.pml for more information."
"IK11","Display architecture for L4","IK","Ihor Kuz","","","Active","All Students","R & D","","Explore the possibilities for a display (GUI) architecture for L4.It should be modular and flexible so that it can be used on many different kindsof (embedded) displays (PDAs, phones, watches, media players, etc.).The project will include exploring existing embedded graphicsarchitectures (e.g., <a href=""http://trolltech.com/products/qt/qt3/embedded/"">QT/embedded</a>) and looking at how concepts fromthese can be applied to L4 and/or improved upon.","","","","See http://www.ertos.nicta.com.au/education/theses.pml for more information."
"IK22","Component Architecture on Secure Microkernel","IK","Ihor Kuz","","","Active","CECS","R & D","","<a href=""http://www.nicta.com.au/research/projects/secure_embedded_l4"">seL4</a> is a new secure version of the L4 microkernel. <a href=""http://www.nicta.com.au/research/projects/component_architectures_for_microkernel-based_embedded_systems"">CAmkES</a> is acomponent architecture designed for building microkernel-basedoperating systems. Currently CAmkES is based on L4 and does notaddress security issues. The aim of this project is to getCAmkES working with seL4 and then explore the ways that seL4 securityfeatures can be leveraged by CAmkES to build secure embedded systems. ","","","","See http://www.ertos.nicta.com.au/education/theses.pml for more information."
"IK23","Shared resources in an L4-based OS","IK","Ihor Kuz","","","Active","CECS","R & D","","One of the key services that an OS provides is a managing access toshared resources.  For example, a file system manages access to shareddisk space, a network stack manages access to a network device, awindow system manages access to the display, etc.  In a modular,microkernel-based OS, these shared resources are managed by user-levelservices.  In this project you will investigate ways of modelling suchshared resource managers within the <a href=""http://www.nicta.com.au/research/projects/component_architectures_for_microkernel-based_embedded_systems"">CAmkES</a> component framework onL4 and develop a suitable model for building such services in acomponentised environment.  You will assess the suitability of thismodel by designing, implementing, and evaluating one or more suchservices (e.g., a file system, a network stack, etc.).","","","","See http://www.ertos.nicta.com.au/education/theses.pml for more information."
"IK24","Non-C languages on L4","IK","Ihor Kuz","","","Active","CECSSE","R & D","","Currently almost all software for L4-based systems is written in C.There are however many languages whose runtimes or interpreters couldbe ported to run on L4 as well.  For example, Python was ported to<a href=""http://www.ertos.nicta.com.au/research/mungi/"">Mungi</a>, an early version of <a href=""http://iolanguage.com/about/"">Io</a> was made to run on L4, as was a verysimple version of the <a href=""http://www.squeak.org/"">Squeak</a> Smalltalk system, and <a href=""http://www.lua.org/"">Lua</a> was also portedto L4/Iguana.  However, none of these run on current versions and, assuch, we are still limited to using C for programming all L4-based applications.  Inthis project you will choose a favourite language andport its runtime or interpreter and critical libraries to run on L4.  Furthermore youshould integrate the language into the L4 environment such that codewritten in that language can invoke L4 system calls, perform IPC andtransfer data to processes implemented in other languages.","","","","see http://www.ertos.nicta.com.au/education/theses.pml for more information.see also GH94: JVM on L4"
"JAS02","4th-Year Thesis Management System","JAS","John Shepherd","","","Active","All Students","Development","","Build a database-backed web-site to manage all aspects of 4th-year theses, from finding topics, to timetabling presentations, to marking theses. You will need to apply sound software engineering methods to this project and produce a well-designed, usable, efficient, extendible, maintainable system *and* demonstrate convincingly that it has these charcteristics.","","COMP3311, COMP3511, solid Web/DB programming skills (including PHP, PostgreSQL, Javascript)","Apache/PHP4/PostgreSQL","If you aren't going to produce a robust, usable system before week 5 of Thesis B, don't apply for this topic."
"JAS10","Online Management of UNSW Course/Program Proposals","JAS","John Shepherd","","","Active","All Students","Development","","Course and Program proposals at UNSW are passedthrough various committees (School teaching committee, Faculty education committee, Academic Board) largely by way of paper (Word) documents and verbal discussion in committee meetings. Proposals also require approval from various authorities in UNSW (e.g. the Library), and the process for this is ad hoc. This project aims to develop a system that can manage the workflow of academic proposals in UNSW, allowing the proposals themselves to be prepared more easily, available to a wider audience than just committees, comments to be collated on line, and approvals to be sought automatically when a proposal passes the School committee. The system must be flexible in allowingthe format of proposals to change over time. It must also provide authentication and levels of authority for who can ""sign off"" (electronically) on proposals.","","COMP3311, COMP3511, solid Web/DB programming skills (including PHP, PostgreSQL, Javascript)","PostgreSQL, PHP, Grieg","If you aren't going to produce a robust, usable system before week 5 of Thesis B, don't apply for this thesis."
"JAS12","CSE Database","JAS","John Shepherd","","","Active","All Students","Development","","Develop a module for the Web-based database which is currently being developed to manage CSE's mission-critical data. Which part of the system you work on is open to negotiation. Note that developing filters to extract information from one source and upload it to the database is one kind of module I'd be interested in. Other kinds of modules would manage specific workflows within CSE. (A specific example of such a module is the topic AN10). You will be expected to produce an extremely usable, robust and scalable system, and demonstrate that it meets these requirements. The system must be built on the Apache/PHP/PostgreSQL platform.","http://www.cse.unsw.edu.au/","COMP3311, COMP3511, solid Web/DB programming skills (including PHP, PostgreSQL, Javascript)","NSS, Research Office, and a host of tables already developed for the CSE DB.","If you aren't going to produce a robust, usable system before week 5 of Thesis B, don't apply for this topic."
"JAS15","Online Roleplay Simulation System","JAS","John Shepherd","","","Active","All Students","Development","","Almost ten years ago, a 4th-year student wrote a very nice web-based system for running on simulation role-plays for Middle East Politics classes at Macquarie University. Web technology has advanced considerably since then. It's time to build a new system that is more flexible, easier to manage, and better-looking than the original system. The system should be able to deal with multiple simultaneous role-plays and deal with any topic area, not just Middle East Politics.","http://mahler.cse.unsw.edu.au/meps/","COMP3311, COMP3511, solid Web/DB programming skills (including PHP, PostgreSQL, Javascript)","Apache/PHP4/PostgreSQL","If you aren't going to produce a robust, usablesystem before week 5 of Thesis B, don't apply forthis topic."
"JAS21","Content-based Image Retrieval Platform for the Web","JAS","John Shepherd","","","Active","All Students","R & D","","Web Sites such as Flickr allows users to describe the semantics of their pictures using keywords (aka tags). Having to tag a large collection of existing images is not feasible, and so this thesis aims to examine ways of automatically classifying images via (some representation of) their content. The specific aim of this project is to assemble a web-based platform for exploring a variety of content-based image retrieval techniques. One goal of the project is to have a compelling demo system available for Open Day 2007.","Go to Google and type ""content based image retrieval""","Serious database and programming skills (many of the existing image analysis tools that you'll be using are written in C and C++)","Existing image analysis packages.PostgreSQL. Web server of your choice.","You'll need to work with feature-extraction packages written in C and C++.If you aren't going to produce a robust, usablesystem before week 5 of Thesis B, don't apply forthis topic."
"JAS22","Content-based music retrieval platform for the Web","JAS","John Shepherd","","","Active","All Students","R & D","","Most music-centric web sites provide access to music using keywords (aka tags) and/or categories. Having to tag/categorise a large collection of existing music is not feasible, and so this thesis aims to examine ways of automatically classifying music data via (some representation of) its content. The specific aim of this project is to assemble a web-based platform for exploring a variety of content-based music retrieval techniques. One goal of the project is to have a compelling demo system available for Open Day 2007.","Go to Google and type ""content based music retrieval""","Serious database and programming skills (many of the existing music analysis tools that you'll be using are written in C and C++)","Existing music analysis packages.PostgreSQL. Web server of your choice.","If you aren't going to produce a robust, usablesystem before week 5 of Thesis B, don't apply forthis topic."
"JAS30","Web-Based Course Management System (WebCMS)","JAS","John Shepherd","","","Active","All Students","Development","","Work on some part of the WebCMS system, which was developed as a 4th-year thesis in 2002 and has been used by over 100 courses to manage their web sites since then. The precise aspect of WebCMS to be tackled is open to negotiation. It currently needs its session system to be reviewed, the privileges system to be generalised, to be made more secure, and to have a better MessageBoard, Group/Project Management and On-line Exercises systems.","","COMP3311, COMP3511, solid Web/DB programming skills (including PHP, PostgreSQL, Javascript)","Apache/PHP4/PostgreSQL","If you aren't going to produce a robust, usable system before week 5 of Thesis B, don't apply for this topic."
"JAS50","UNSW Library","JAS","John Shepherd","","","Active","CECSSE","Development","","The library has a number of development projects that they'd like CSE thesis students to work on. These include: using PDAs to assist library staff to find relevant information; developing a room booking service; interactive web services (e.g. getting directions to the shelf a book is on once you have located it in the catalogue); extracting and analysing traffic patterns of web service usage. The precise topic would be negotiated between the student, the supervisor and library staff.","","COMP3311, COMP3511, solid Web/DB programming skills (including PHP, PostgreSQL, Javascript)","","If you aren't going to produce a robust, usablesystem before week 5 of Thesis B, don't apply forthis topic."
"JAS51","Personal Information Management","JAS","John Shepherd","","","Active","BINFBIOMCECSSE","Research","","Personal Information Management deals with the organisation of personal data (emails, documents, calendar events, pictures, itineraries, etc.) into a coherent data space. At present this data space is fractured across application boundaries (e.g. your email is handled by one application, your documents by another). This project aims to investigate how such problems might be overcome.","","Background in databases and (optionally) XML and meta-data.","",""
"JAZ01","Night-time Vehicle Detection and Tracking in Surveillance Video","JAZ","Jian Zhang","","","Active","All Students","Development","","Advanced traffic monitoring system, through automatic analysis of traffic surveillance video, can provide information to significantly improve traffic management and reduce traffic delay and incidents. The Multimedia and Video Communication (MMVC) team in NICTA has been working closely with the NSWÕs Road and Traffic Authority to develop such an intelligent system. The goal of this project is to online detect and track moving vehicles captured by a road-side camera at night. The tasks of this project include understanding the existing nighttime vehicle detection and tracking methods combined with NICTAÕs approach and developing test software of algorithms to test  video sequences captured in various conditions. This is a challenging and exciting project. You will have access to the state-of-art video processing tools and facilities. You will find that the knowledge which you have learned from your courses will be extremely useful in this project. Your programming skills will also be greatly enhanced by doing this project.","[1] Zehang Sun, George Bebis, and Ronald Miller, ÒOn-Road Vehicle Detection: A ReviewÓ, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 28, No. 5, May 2006[2] R. Cucchiara, M. Piccardi, ÒVehicle Detection under Day and Night IlluminationÓ, Proceeding of ISCS-IIA 1999, Special Session on Vehicle Traffic and Surveillance.. [3] R. Taktak. M. Dufaut and R. Husson, ÒVehicle Detection at Night Using Image Processing and Pattern Recognition, IEEE International Conference on Image Processing, 1994.","C/C++, Advanced Math, and course subjects mentioned in the project description","",""
"JAZ02","Multi-sensor Data Fusion for Real Time Traffic Monitoring  ","JAZ","Jian Zhang","","","Active","All Students","R & D","","Sensor data fusion is to combine data derived from disparate sources. Such resulting information is expected to have more rich background signals from different ÒanglesÓ which in some sense is better than would be possible when the information source  were  used individually. For real time traffic monitoring, it is expected to have  a set of sensors such as video camera, radar, intra-red cameras, optical cameras and leaser signal sensors to capture traffic information in different signal formats and provide multi-signal dimensions to achieve traffic event detection and analysis. The data fusion is based on the signal sampling from multi-sensors and algorithm development  in the domain of statistic signal processing  such as Kalman filter and Bayesian Networks. The knowledge you learned from the computer and engineering courses, such as advanced math, signal processing, and statistics theory will gain plenty practice from this project.  Your programming skills will also be developed as you will develop a demo in software for data capturing from the different sensors and simple fusion algorithm implementation. ","[1]IEEE Sensors Journal[2] Proceedings of The 2005 International Conference on Data Fusion - From Multi-Source Data to Information","Advanced programming skill in C++ and Java","",""
"JAZ03","Objects Classification and Event Detection in Surveillance Video.","JAZ","Jian Zhang","","","Active","All Students","R & D","","Object classification is a further step to object detection and tracking in understanding of video. Without object classification, we can only know something is moving in the video sequences, but donÕt know what it is. With the information from object tracking and classification, it may become possible to understand the activities and events happening in the video sequences. The task of this project is to investigate the existing object classification technologies and develop one for categorizing moving vehicles or human detected from the surveillance videos. This is a challenging, but also exciting project. You have access understate the state-of-art video processing tools and facilites. The knowledge you learned from the courses of computer vision, neural network, pattern recognition and machine learning can all find their utilization in this project. Your programming skills will also be developed as you will develop a demo in software","Reference:[1] T. Kanade, R. Collins, A. Lipton, P. Burt, and L. Wixson, ""Advances in Cooperative Multi-Sensor Video Surveillance"",Darpa Image Understanding Workshop, Morgan Kaufmann, November, 1998, pp. 3-24 [2]R. Collins, A. Lipton, H. Fujiyoshi, and T. Kanade, ""Algorithms for cooperative multisensor surveillance"", Proceedings of the IEEE, Vol. 89, No. 10, October, 2001, pp. 1456 - 1477. [3] Wei Niu, Jiao Long, Dan Han and Yuan-Fang Wang, ÒHuman Activity Detection and Recognition for Video SurveillanceÓ, ICME 2004[4] Surendra Gupte, Osama Masoud, Robert F. K. Martin, and Nikolaos P. Papanikolopoulos, ÒDetection and Classification of VehiclesÓ, IEEE Transactions on Intelligent Transportation Systems, VOL. 3, NO. 1, December 2002","C/C++ and Java programing, Advanced Math","",""
"JAZ04","Robust Background and Foreground Analysis for Detecting and Tracking Moving Objects in Surveillance Video.","JAZ","Jian Zhang","","","Active","All Students","Development","","The ultimate goal of any video surveillance system is to automatically understand and detect events of interest happening at a monitored site. Such a high-level task requires some low-level computer vision tasks to supply necessary information. The goal of this project is to utilize a recently proposed nonparametric Kernel Density Estimation (KDE) based algorithm to analyze the background and foreground to robustly detect moving objects out of dynamic environment and track their movement under occlusion situations. The knowledge you learned from the computer and engineering courses, such as computer vision, signal processing, and statistics theory will gain plenty practice from this project.  Your programming skills will also be developed as you will develop a demo in software.","[1] Ahmed Elgammal, Ramani Duraiswami, David Harwood, and Larry S. Davis, ÒBackground and Foreground Modeling Using Nonparametric Kernel Density Estimation for Visual SurveillanceÓ, Proceedings of The IEEE, Vol. 90, No. 7, July 2002. ","C++ programming. Advanced Math","",""
"JAZ05","A Real-time Motion Detection and background modeling for Video Surveillance Systems","JAZ","Jian Zhang","","","Active","All Students","R & D","","The advance of fast and affordable computing device has provided the power to track moving objects from a video camera in real-time. Many algorithms have been developed for motion detection and background modeling in the past, using various image/video features, such edge, color, spatial-temporal information, and statistical model. The task of this project is to investigate the existing background modeling developed in the Multimedia and Video Communication research group in NICTA Sydney Kensington Lab. The task will support the researcher to develop robust algorithms for a real time surveillance video system. The knowledge you learned the courses from NUSW such as multimedia processing and computer vision will gain plenty practice from this project. Your programming skills will also be developed as you will develop a demo in software.","[1] T. Kanade, R. Collins, A. Lipton, P. Burt, and L. Wixson, ""Advances in Cooperative Multi-Sensor Video Surveillance"",Darpa Image Understanding Workshop, Morgan Kaufmann, November, 1998, pp. 3-24[2]R. Collins, A. Lipton, H. Fujiyoshi, and T. Kanade, ""Algorithms for cooperative multisensor surveillance"", Proceedings of the IEEE, Vol. 89, No. 10, October, 2001, pp. 1456 - 1477. ","C++ programming and advanced Math (undergraduate level)","",""
"JAZ06","Pedestrian Detection Using a Cascade of Boosted Classifiers","JAZ","Jian Zhang","","","Active","All Students","R & D","","This project aims provide an implementation, verification and evaluation of a pedestrian detection framework proposed by Viola-Jones which is fast and robust under varying conditions by incorporating both motion and appearance in training a cascade classifiers. It is hoped that this project will serve as a foundation for other applications such as vehicle and cyclist detection; and incorporated with other image processing techniques such as tracking which would improve robustness and computational efficiency.","Paul Viola, Michael J. Jones, Daniel Snow. ""Detecting Pedestrians using Patterns of Motion and Appearance"", ICCV 2003. November 27http://www.merl.com/papers/docs/TR2003-90.pdf","Digital image processing, Computer vision, C++ programming and Advanced Math (undergraduate level, Matlab.","",""
"JAZ07","Semantic-Sensitive Image Classification","JAZ","Jian Zhang","","","Active","All Students","Development","","Due to the rapidly growing amount of digital image data on the Internet and in digital libraries, there is a great need for large image database management and effective image classification tools. Image classification algorithms typically attempt to label an image as either true or false based on the presence or absence of some target concept with semantic meaning, or assign an image into one of several semantic categories. Potential applications of image classification range from organizing a personal photo collection to detecting specific objects within a surveillance system. This project is focused on improving and enhancing the image classification system developed by Multimedia and Video Communication Group at NICTA. The main tasks involved in this project are: (1) implementing an image segmentation technique and integrating it into the image classification system; (2) modifying some implementations and improving the efficiency of the system. You will have an opportunity to learn how to use image processing, image retrieval, and machine learning techniques to solve several critical problems.","[1] Stepehn Pek, ÒA stacked ensemble of image classifiers using a per-image weighting schemeÓ, Thesis, School of Information Technologies, The University of Sydney, 2005.[2] J. Wang, J. Li, and G. Wiederhold, ÒSIMPLIcity: Semantics-sensitive integrated matching for picture librariesÓ, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 23, No. 9, 2001[3] J. Bi, Y. Chen, and J. Wang, ÒA sparse support vector machine approach to region-based image categorizationÓ, IEEE International Conference on Computer Vision and Pattern Recognition, 2005.","C/C++, Advanced Math","",""
"JAZ08","A Real-time Object Tracking System for the Video Surveillance","JAZ","Jian Zhang","","","Active","All Students","R & D","","The advance of fast and affordable computing device has provided the power to track moving objects from a video camera in real-time. Many algorithms have been developed for object tracking in the past, using various image/video features, such as edge, color, spatial-temporal information, or using Kalman filtering and hypothesis testing techniques. Robustly tracking object under the occluded situation is one of the challenging topics in this area. The task of this project is to investigate ","[1] T. Kanade, R. Collins, A. Lipton, P. Burt, and L. Wixson, ""Advances in Cooperative Multi-Sensor Video Surveillance"",Darpa Image Understanding Workshop, Morgan Kaufmann, November, 1998, pp. 3-24                                             [2]R. Collins, A. Lipton, H. Fujiyoshi, and T. Kanade, ""Algorithms for cooperative multisensor surveillance"", Proceedings of the IEEE, Vol. 89, No. 10, October, 2001, pp. 1456 - 1477. [3] Tao Xiong and Christian Debrunner, ÒStochastic Car Tracking With Line- and Color-Based FeaturesÓ, IEEE Transactions on Intelligent Transportation Systems, VOL. 5, NO. 4, December 2004[4] Yue Zhou and Hai Tao, ÒA Background Layer Model for Object Tracking through OcclussionÓ, ICCV 2003","C++ programming ","",""
"JAZ09","Video coding and transmission over wireless networks","JAZ","Jian Zhang","","","Active","All Students","R & D","","It will be the part of our Video Analysis and Content Management research project. Wireless video/image communication will bring huge multimedia applications. As a strong demand from mobile and PDA industry, the research and R/D are moving fast.  The video coding and its communication protocol have been developed by the MPEG/ITU-T standard organizations and Internet Engineering Task Force (IETF). There are lots of potential video coding and image processing algorithms to achieve high process performance with low power consumption.  The task for this project will require some C/C++ coding skills and some courses training including multimedia technology, advanced Math, software development and image processing. One possible research area is in the content based video coding that was defined in MEPG-4 standard.  This research could be extended to a postgraduate study. ","[1] Barry G. Haskell, Atul Puri & Arun Netravali, ""Digital Video -- An Introduction to MPEG-2"", Kluwer Academic Publishers 1999","Digital image processing, digital signal processing, C++ programming and Advanced Math (undergraduate level, Matlab.","",""
"JAZ10","Content-based image retrieval with support vector machines","JAZ","Jian Zhang","","","Active","All Students","R & D","","Feature selection has been a preliminary step to preprocess high dimensional data for mining, indexing and visualization. Generally, functionality of feature selection includes two major components 1) dimension reduction Ð effectively reducing size of raw feature size, 2) optimal feature subset selection. In this project, we mainly focus on large multimedia data. An advance feature selection scheme will be developed to facilitate efficient and effective classification of large multimedia dataset and compared with state of art approaches.","Tom Mitchell, Machine Learning. McGraw-Hill, 1997.B. Prabhakaran, Multimedia Database Management Systems, Kluwer Academic Publishers 1997","Advanced programming skill in C++ and Java","",""
"JAZ11","Video Content Database Management on Server ","JAZ","Jian Zhang","","","Active","All Students","Development","","The vide storage server will store the video sequence received by the server. To support the efficient retrieval, the server shall support following functions: 1. create a XML file to describe the video data being stored; 2. add time stamp information to the XML file, 3. obtain the key frame information (I frame & short boundary info) from the video sequence and insert them to the XML file, 4. Send the generated XML file to a central server for storage. 5. able to get a XML file from the central server, analyse the metadata in the XML file, and retrieve the corresponding video sequence from the video storage server. This project will be carried out in multimedia & video communication research group in National ICT Australia (NICTA) laboratories Ð Sydney Kensington Lab. You will work directly with many talent researchers in a high standard image/video lab and will be given clear guidance for work that could link to your future career development. After this project, you will have good experience and knowledge at video coding, video communication, TCP/IP and Http protocol, and Native XML database.","","C++ programming, Advanced Math (undergarduate level)","",""
"JAZ12","Semantic Video Understanding Based on HMM framework","JAZ","Jian Zhang","","","Active","All Students","R & D","","Content based video analysis has attracted my research groups to explore the semantic representation for multimedia content. There are several key low feature descriptions defined in MPEG-7 standard with its descriptors. One challenge to multimedia research is how to use these low level features to achieve some semantic understanding of the video sequence. This will involve the low-level feature extraction, Gaussian model expression, and HMM model training and testing, which will support to generate the semantic understanding of a video sequence automatically. The knowledge you learned from the courses of computer vision, pattern recognition and multimedia technology such as image/video processing can all find their utilization in this project. Your programming skills will also be developed as you will develop a demo in software.  ","[1] L. Xie, S.-F. Chang, A. Divakaran and H. Sun (2003), Feature Selection for Unsupervised Discovery of Statistical Temporal Structures in Video, IEEE International Conference on Image Processing (ICIP 2003), Barcelona, Spain, September 2003.[2] L. Xie, S.-F. Chang, A. Divakaran and H. Sun (2003), Unsupervised Discovery of Multilevel Statistical Video Structures Using Hierarchical Hidden Markov Models, ICME 2003, Baltimore, MD, July 2003","","C++, Java and some knowledge of XML programming",""
"JAZ14","Steraming video over wireless networks","JAZ","Jian Zhang","","","Active","All Students","R & D","","Although current wireless mesh networks are still mainly used for data transmission, the demand for video transmission has been increased in many applications. For this project, you will build up a testbed for video over wireless networks. You will be offered a great chance to do research to develop video transmission protocol and error protection scheme.    ","Danjue Li, Chen-Nee Chuah, Gene Cheung and S.J. Ben Yoo ""MUVIS: Multi-source video streaming service over WLANs"", in Journal of Communication and Networks (JCN) 2005","Networking protocol, singal processing and wireless networksC++, ","",""
"JAZ15","3D video/image processing","JAZ","Jian Zhang","","","Active","All Students","Development","","There are challenges opportunities in developing and deploying 3D TV services. The 3D TV services can be seen as a general case of the multi-view video that has been receiving significant attention lately. Thinking about improvement of the 3D TV from the view of video encoding, there are appoaches such as the combination of view interpolation (using a per-pixel depth map or other improved computer vision model) and view synthesis prediction or some improved approaches for redundancy removing. The task for this project will require some C/C++ coding skills and some courses training including Computer Vison, Advanced Math, software development and image processing. This thesis topic could be extended to a postgraduate study. ","[1] Barry G. Haskell, Atul Puri & Arun Netravali, ""Digital Video -- An Introduction to MPEG-2"", Kluwer Academic Publishers 1999","C++ programming","",""
"JIC08","Automatic Vision-Based Incident Detection","JIC","Jing Chen","","","Active","All StudentsBINF Priority","Research","","Due to the ever increasing traffic density for most major intersections and highways, automatic incident detection (AID) has become a main source for the Traffic Management Centre (TMC) to schedule the traffic control and further alleviate the traffic congestion. Using vision-based AID systems to obtain real-time incident information automatically and precisely has attract significant research focus recently. In the project, you will investigate the state-of-the-art technologies being applied in the AID system that includes three consecutive steps: object detection, vehicle tracking and activity understanding, and to further compare the performance of those algorithms. You will be using those latest technologies and tools in areas of computer vision and machine learning etc. A good C/C++ programming skill will be highly appreciated.","","C/C++","",""
"JIC09","Video Communication Over Wireless Networks","JIC","Jing Chen","","","Active","All Students","R & D","","Sending video sequence over wireless network where retransmission is infeasible or too expensive is a hot topic for video communication over wireless networks. Cross-Layer coding has been considered as a promising solution to offer error resilience in video communication especially under varying and limited channel bandwidth situation. In this project, you will study the architecture of the cross-layer video coding, investigate kinds of existing cross-layer coding technologies for a wireless network, and finally setup a simulation platform to evaluate the performance of your schemes. After this project, you will gain some solid knowledge and experience at video coding, TCP/IP, and the protocol of networking player of a wireless network.","[1] Yan Wang, ÒMultiple Description Coding for Video DeliveryÓ, PROCEEDINGS OF THE IEEE, VOL. 93, NO. 1, JANUARY 2005[2] V. K. Goyal, ÒMultiple description coding: Compression meets the network,Ó IEEE Signal Processing Mag., vol. 18, pp. 74Ð93, Sep. 2001.","C/C++","",""
"JOT05","Patient mean used as a Quality Control measure","JOT","John Teal","","","Active","BINFBIOM","Research","","Medical diagnostics are performed by automated machines that measure a wide range of biological indicators such biochemistry, haematology and immuonlogy tests. At regular intervals QC tests are performed on standard materials to validate the correct operation of the machine. It is believed, but not proven that an average or other statistcial measurement of patient results can supplement the operation of normal QC tests.This project requires a student who has an interest in basic statistics to review existing literature on the subject. The student will have access to a relavant scientist at a choice of major Sydney hospitals.Optionally, the student can be provided with actual patient results that could be used for processing on a spreadsheet or statisical package.Also optionally, the student will be given access to an Australian product, developed in Microsoft .NET technology that is in development to store and display patient results.The core of this project is to research the topic to assess relevant progress in this field. If appropriate, the student has the opportunity to test various models on a spreadhseet. And if a clear direction is found to make recommendations for the further development of an Australian product.The student will be offered technical assistance and project guidance from a specialist software developer based in Sydney.","","","","Basic statistics. Basic biometric test usage. Spreadsheet technology."
"JP10","Intensional Semantics of Programming Languages","JP","John Plaice","","","Active","CECSSE","Research","","The TransLucid language, designed by John Plaice, is used by programmers to define variables as infinite, multidimensional entities.  An interpreter exists for the language, and it has been designed so that it can be used as the implementation language for other programming paradigms.The proposed research consists of defining means for translating functional, logic and object-oriented programs into TransLucid, and doing so in a transparent manner, in which the original program is recognisable in the output.  The translations will be proven correct, implemented, and the results will be published.This topic consists of both theoretical and practical work, and is of interest to students considering a research career in programming language semantics or implementation.","","","","Contact John Plaice for further information."
"JP11","Novel Implementations of TransLucid","JP","John Plaice","","","Active","CECSSE","Research","","The TransLucid language, designed by John Plaice, is used by programmers to define variables as infinite, multidimensional entities.  An interpreter exists for the language, and it has been designed so that it can be used as the implementation language for other programming paradigms.The proposed research consists of developing highly efficient implementations for TransLucid that allow it to be used in regular programming situations.This topic consists of both theoretical and practical work, and is of interest to students considering a research career in programming language semantics or implementation.","","","","Contact John Plaice for further information."
"JP12","Cartesian Programming of Distributed Systems","JP","John Plaice","","","Active","CECSSE","Research","","It is increasingly recognised that the standard models of programming distributed and timed systems are too ad-hoc, and do not allow general development techniques, independent of the underlying hardware architecture.In this thesis, the student is expected to generalise existing Cartesian programming techniques, as exemplified by the TransLucid language, to support the development of distributed and timed systems.The student will be taught about programming language semantics, for sequential and distributed systems, and will develop new primitives for distributed programming.","","","","Please contact John Plaice for further details."
"JYL",".Net (or Java) Event Driven Architecture for Real-Time Finanacial Applications ","JYL","Jenny (Yan) Liu","","","Active","CSSE","R & D","","Complex Event Processing (CEP) is the technology to process events and discover complex patterns among multiple streams of event data. CEP has broad applications in different domains, such as money exchange in financial applications. This project aims to design and develop an event driven architecture for foreigh exhange with high demand on real-time and dependability features. .Net platform is adopted (Java platform might also be considered) to realize this architecture. An open source  event stream processing infrastructure Esper.Net (or Esper Java) is used as the programming model. Applicants of this project should be motivated by learning new/advanced technologies. .Net programming experience (or server-based Java programming skills) is required.  ","http://esper.codehaus.org/","","",""
"JYL00","Developing a Service Oriented Architecture based on OSGi Framework for Networked Architecture","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","This project aims to develop a service oriented device architecture by leveraging OSGI framework. OSGI provides the platform specification for seamlessly connect devices with services such as REST/http or SOAP-based web services. The architecture is motivated by a vineyard management system, managing communication between desktop based information management system, PDA access and sensors. Simulation techniques are applied to mimic the deployment of a vineyard envrionment. Students are expected to pratice their software development skill from requirement analysis, architecture design, implementation and testing. Students can also gain experience with advanced middleware technology of OSGI. ","","","",""
"JYL01","Developing a web server on mobile phone using OSGI framework","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","With the increasing hardware computing capacity of devices such as mobile phones, deploying a web server on mobile phone is feasible and practical. With a web server embedded in a mobile phone, web sites can be hosted on a mobile phone, personal web information such as a blog, home page, bookmarks can be updated and published anywhere anytime with the mobile phone at hand. More interestingly, this can enable users to access their mobile phone to get call history, missed calls etc, even if the phone was left at home. One way to implement such a web server on devices is leveraging OSGI framework and specification. OSGI aims to seamlessly connect devices and services. The nature of this work is a pilot project, focusing on solving the web server element design and implementation as OSGI components and porting to specific mobile phone devices.  The testbed of this implemetation is through simulation software. Students will gain tremendous experience with web services, web technology and mobile phone based development.  ","","","",""
"JYL02",".NET-based Auction Service Design and Evaluation: managing trustworthiness","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","",".Net platform provides rich service support for developing systems to address security requirements. Evaluating .Net security architecture can provide insights in developing trustworthy applications. An example application is modeled and implemented by simulating the eBay auction service on .Net. Alternative .Net security architecture design patterns are identified and evaluated according to this auction service. Both quantitative and qualitative evaluation results are presented at the end of this project. Students are exposed to various .Net technologies and web services. ","","","",""
"JYL03","Enhance the adaptivity of Integrated Services in SOA","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","Developing adaptive middleware platforms is in the spirit of autonomic computing for building systems with self-managing, self-configuring and self-optimizing behavior. It includes components and services that facilitate the development of adaptive server-side applications. NICTA's Adaptive Middleware Platform project has developed a prototype for J2EE based adaptive server framework (ASF), which defines a component model for primary type of adaptive control components and their communication patterns. Currently the prototype can be deployed on JBoss and WebLogic application server. Monitoring and controlling the states of ASF components are essential to trace and manage the runtime behavior of these components.  Currently ASF components can only be interacted by through programmed code. The aim of this project is to develop a management console as a web portal to ASF at runtime. We apply MVC approach to develop web interface for access to the ASF components. This includes the design and implementation of a management layer at ASF to interact with ASF components through their well defined interfaces and a web portal application to interact with this management layer. The implementation will require web technologies including Ajax, Struts, Apache Tomcat Web Server and JBoss Application Server.  Student can learn and apply various leading web technologies. This project is R&D oriented with 70% programming effort estimated. The student will work closely with NICTA researchers for both academic supervision and technical support. The workspace and machines can be arranged by NICTA at Australian Technology Park if desired. LINKS TO IMPORTANT WEBSITES:http://www.cse.unsw.edu.au/~yliu/tor.htm http://nicta.com.au/director/research/projects/m_to_r/meaamp.cfm SKILL PREREQUISITES:Students are expected to have strong Java programming skills. Experience with web application development is required. Minimal WAM 70% for undergraduate. For more information about the Adaptive Middleware Platforms project, please visit http://nicta.com.au/director/research/programs/ese/research_projects.cfm Contact: jenny.liu@nicta.com.au ","","Students are expected to have strong Java programming skills. Experience with web application development is required. Minimal WAM 70% for undergraduate.","",""
"JYL04","Performance Modelling of Enterprise Applications on Service Bus","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","Performance Modelling of Enterprise Applications on Service Bus","","","",""
"JYL06","The model and architecture of adaptive components on .Net platform","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","Adaptive application server technology is in the spirit of IBM'sautonomic computing initiative for building systems with self-managing,self-configuration and self-optimization behaviors. In such a system,an application server acts as an autonomic element which executes in a dynamic and ever-changing environment. This requires the applicationserver to be enabled to adapt to the environmental changes at runtime.It is a challenging research issue to develop efficient infrastructure support to achieve this goal, which has potential and practical impact on the current middleware technologies, such as J2EE and Web Services.In NICTA's ESE program, we have developed an adaptive server framework thatcan extend application servers with adaptive behaviors without changing the source code of either the existing applications or the application server itself. This is anon-going project and several sub-projects are derived from it. The aim of this project is to design and implement a web portal for the adaptive server framework, so that the management tasks and runtime status information can bedone from a web-based rich user interface. We consider to apply Ajax technology sothat the web portal can be dynamically and automatically updated through asynchronous communication model supported by Ajax. The project development will use ApacheMaven and follow its project management process. ","","","",""
"JYL07","Monitoring Adaptive Server Applications using Eclipse Test and Performance Tools Platform","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","Adaptive application server technology is in the spirit of IBM'sautonomic computing initiative for building systems with self-managing,self-configuration and self-optimization behaviors. Developing adaptive server applicationsrequires comprehensive infrastructure supports including flexible communication models,efficient tools for monitoring runtime execution and management services to coordinate themultiple adaptive behaviors. In NICTA's ESE program, we have developed an adaptive server framework thatcan extend application servers with adaptive behaviors with multiple external control components without changing the source code of either the existing applications or the application server itself. The adaptive behavior is implement inside the control components supported the the adaptive server framework.The aim of this project is to develop a monitoring toolkit for the adaptive server framework that can monitor and visualize the runtime status of control components. This work will utilizeEclipse Test and Performance Tools Platform (TPTP) and develop instrument API for  the current adaptive server framework implementation.  TPTP provides comprehensive frameworks  for building monitoring tools of application servers (JBoss, Jonas, and Websphere) and system performance.It also support collecting, analyzing, aggregating, and visualizing data captured in the log and statistical models.The project will use Apache Maven as the build manager.","","","",""
"JYL08","Performance evaluation of architectural alternatives of .NET Applications","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","We are seeking three to four students to work on performance evaluation of .Net architectures. The work will involve building several multi-tier .NET applications and executing benchmarks to understand the performance of .NET applications. Each student will work on one of the followingaspects: .Net webservices, .Net remoting, messaging with MSMQ and .Net, security management, transaction control, data access in ADO.Net. Exposure to various technologies such as Web Services, various .NET languages and SQL Server will be necessary. Experience will also be gained with methods and tools for performance measurement and profiling.","","","",""
"JYL09","Porting Adaptive Server Framework on COTS Application Servers","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","Adaptive application server technology is in the spirit of IBM'sautonomic computing initiative for building systems with self-managing,self-configuration and self-optimization behaviors. Developing adaptive server applicationsrequires comprehensive infrastructure supports including flexible communication models,efficient tools for monitoring runtime execution and management services to coordinate themultiple adaptive behaviors. In NICTA's ESE program, we have developed an adaptive server framework thatcan extend application servers with adaptive behaviors with multiple external control components without changing the source code of either the existing applications or the application server itself. The adaptive behavior is implement inside the control components supported the the adaptive server framework.This project aims to port the current adaptive server framework implemented in Java on COTS application servers including IBM WebSphere and BEA WebLogic Server. ","","","",""
"JYL11","Model Driven Development for Adaptive Middleware-based Applications: An Eclipse plugin","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","Autonomic computing is of IBM's initiative for building systems with self-managing,self-configuration and self-optimization behaviors. Autonomic computing is a promising way to enable middleware based systems to adapt to rapid and continuous changes of the environment.  Developing adaptive middleware-based applications requires comprehensive infrastructure supports including flexible communication models, efficient tools for monitoring runtime execution and management services to coordinate the multiple adaptive behaviors. In NICTA's ESE program, we have developed an adaptive server framework thatcan extend application servers with adaptive behaviors with mulitple external control components without changing the source code of either the existing applications or the application server itself. The adaptive behavior is implemented inside the control components supported the the adaptive server framework.The aim of this project is to follow Model Driven Development paradigm and build a tool to facilitate the the development of middleware-based applications using our adaptive server framework. This project includesthe design of structure and functionality of the MDD tool that can formally capture the component models,policy specification, service descriptions of adaptive server framework into UML design diagrams, and implementation of the code generation from UML design descriptions to Java code. This work can significantly reduce the programming overhead of developing adaptive server applications, and thus is of great practical values. ","","","",""
"JYL12","Model Driven Development for Microkernel-based Embedded and Real-time Applications","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","Embedded software applications are typically more difficult to design and build, due to the strict resource constraints of the target environmentand the problem domains. Model Driven Development (MDD) is one of the enabling techniques for rapid development of embedded and real-timeapplications.  The principle of model driven development (MDD) is transforming business domain models into Platform Independent Models (PIM) and then Platform Specific Models (PSM), and eventually to executable code. Sharing and working with models on a software development is more efficient and less error-prone than working only in source code. This greatly reduces the programming effort and expertise needed for developing embedded software systems.In NICTA, a joint project CAmkES between ESE and ERTOS program has established a component architecture to supportthe development of embedded systems on top of microkernels. The architecture provides a component model, standard interfaces andcomponent definitions, component implementations, standard services, and support for various architectural patterns suited to embedded systems.The aim of this project is to follow MDD approach and develop a tool that can take (visual) models at a high level of abstraction and generate executable and efficient source code for CAmkES-based applications. This project will involve activities of specifying the semantics of model elements to provide rich information on the structure and behaviour of the system and implementing code generation from the model descriptions. This project will utilize Eclipse Modeling Framework (EMF) and produce the tool as an Eclipse plugin. ","","","",""
"JYL15","Developing loan application on Enterprise Service Bus and the performance evaluation","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","This project will implement a loan application described in the book of Enteprise Integration Pattern on Enterprise Service Bus implemetations including BEA's Service Bus and open source Mule. A performance evaluation will be carried out on this loan implmentation. ","","","",""
"JYL16","Architecture analysis of TinyOS","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","TinyOS is open source OS for developing embedded and sensor network applicatons. Understanding TinyOS's architecture helps to design and implement applications on top of it. This work involves reverse engineering, source code analysis, design pattern identification and survey of the usage of TinyOS","","","",""
"JYL17","Smart endpoint design for simplified middleware","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","This is a very challenging project. Integrating smartness into the endpoint of service based application can signficantly improve the flexiblity and extensibility of service applications. An example is Axis 2 project which provides very flexible configration of the modules in the endpoint. This project aims to design the smart endpoint that can be integrated with Mule service bus platform and provide flexible and lightweight modules for adaptivness.","","","",""
"JYL18","Sharing and Broadcasting Google Maps using Enterprise Service Bus","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","Google maps are very popular and widely used by individual users to find map information quite effeciently. However sharing the map information are now done by copying/pasting links of google maps through messagers or emails. It will be very convinient if google map information can be shared between users and the map information can be simulatously updated when any user manipulates the map. This can be achieved using enterprise service bus, which is an event driven middleware supporting event tranfering between heterogenous platforms and communication protocols. This project aims to enable the google map sharing and broadcasting using enterprise service bus. At the end of the project, a demonstration is expected showing multiple users can share the same google map information with real-time update. ","","","",""
"JYL19","Business-Driven Autonomic Service-Oriented Middleware","JYL","Jenny (Yan) Liu","","","Active","CECSSE","R & D","","   Web services are distributed computing application components that use a number of Extensible Markup Language (XML) based technologies to implement the service-oriented architecture (SOA). Implementation-independence of Web services technologies allows different businesses to collaborate and achieve common business goals despite the fact that the collaborating Web services can be distributed over the Internet, run on different platforms, and implemented in different programming languages. Web services technologies are already embedded in various products and services of all major computing companies and used for diverse purposes.    Management of IT systems (including service-oriented systems) is the process of their monitoring and control to ensure correct operation, discover and fix problems (e.g., faults, performance problems), and achieve maximal benefits from the IT systems. Ideally, IT systems should be self-managing, or at least be managed with minimal human intervention. This is one of the goals of autonomic computing. The current management solutions for service-oriented systems focus on optimizing technical quality of service (QoS, e.g., response time and availability), but provide inadequate treatment for optimizing business metrics, such as profit and customer retention.    Students working on this thesis project will design, implement, and test extensions of an existing autonomic service-oriented middleware (already measuring technical quality of service) with additional software support (implementing new algorithms, data structures, XML formats) for maximization of business metrics. The project will address not only financial business metrics (e.g. profit), but also other measures of long-term business health (e.g., customer retention). In this way, it will overcome limitations of the previous approaches. The expected outcomes have both theoretical and practical character. Theoretical contributions include new XML formats, data structures, as well as algorithms for representing, storing, and processing business information for management of service-oriented systems. Practical contributions include running code that implements and supports the above XML formats, data structures, algorithms, in a way compatible with the existing software. Since businesses ultimately want maximization of business, not technical metrics, this work will have significant practical applications. ","","","","The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Jenny Liu or Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"JYL20","Composing Self-ManagingService-Oriented Systems : A process driven framework","JYL","Jenny (Yan) Liu","","","Active","CSSE","R & D","","Loose coupling and preserving safety are two key criteria for composing self-managing services. Composition of adaptive control components with business services without interfering with the original service operations is complicatedby the dynamic and highly distributed nature of service oriented systems. Essentially, encapsulating control logicinto abstract logical models enables a clear separation ofconcerns, with states and transitions indicating the logicalcontrol flow. The challenge is to seamlessly integrate thesemodels with services and their host infrastructure as a unifiedself-managing environment. In this present, we aims to devise a process driven framework for the composition andcoordination of self-managing services. This architecture framework leverages multidisciplinary  approaches with business process models, autonomic computing, and mechanism designs to producedeclarative and executable control models. ","","Strong programming skills, and interests in research","",""
"KE01","Proving theorems in and about epistemic logics in Isabelle and PVS","KE","Kai Engelhardt","","","Active","CECSSE","Research","","The aim of this project is to build two competing proof assistant tools: one based on the theorem proving environment Isabelle (jointly developed by Cambridge and Munich Universities) and one based on the higher-order logic theorem prover PVS (developed by SRI in Stanford). Each of the proof assistants should help users in proving theorems in and about various epistemic logics (logics of knowledge or belief). It is currently unclear which theorem prover should be preferred for this application and why. A subgoal of this project is to evaluate the two suggested platforms for their respective suitability for the task at hand. One particularly interesting candidate logic would be the logic of local propositions (LLP) for multiple agents. LLP is the basis of an assertion language in a new refinement calculi currently developed. The (better one of the) proof assistants would form one corner stone of a future refinement assistant (similar to the B-Tool).","Isabelle: <http://www.cl.cam.ac.uk/Research/HVG/Isabelle/> PVS: <http://pvs.csl.sri.com> Epistemic logics: <http://www.britannica.com/eb/article?eu=119909&tocid=65893#65893.toc> LLP: <http://www.cse.unsw.edu.au/~kaie/Papers/Ebib/node7.html>","good understanding of mathematical logics","both Isabelle and PVS are (relatively) freely available and run on linux and Mac OS X","This project is advertised for a group of 2 students. It is possible to split it into two almost independent projects. The only mandatory collaboration would consist of writing a joint assessment of the merits not of their proof assistants but of the underlying theorem proving environments (Isabelle and PVS)."
"KE07","Disecting Refactoring","KE","Kai Engelhardt","","","Active","CECSSE","Research","","Refactoring is one of the newer fads of ad hoc software engineering. It comprises of an ever-growing number of allegedly behaviour-preserving program transformation rules. Your task is to investigate one or more of:1. the soundness of these rules2. the effect of refactoring on software metrics scores3. the feasibility of an automated tool that combines genetic algorithms with refactoring and software metrics","http://www.refactoring.com/http://www.eclipse.org/","","","1. is a mildly theoretical topic requiring a modest amount of mathemtical maturity, 2. is empirical in nature, 3. is a buzzword-compliant practical topic perhaps suitable for a group"
"KE09","CEGAR testbed in Haskell","KE","Kai Engelhardt","","","Active","CECSSE","R & D","","CEGAR means Counter-Example guided Abstraction Refinement and refers to one of the most useful techniques to extend the range of problems amenable to model checking. More often than not, real systems and real properties are too complex to be checked against each other automatically in a brute-force manner. It is however possible to check safe abstraction of real systems against mathcing abstraction of properties in a way that guarantees that if the abstraction has the abstracted property then the real system complies with the real property (but typically not vice versa). While a number of industrial strength model checkers are available under open source licenses, there is a distinct lack of a CEGAR testbed in the open that allows researchers to evaluate new abstractions against test suites of systems and properties. The aim of this thesis project is to develop a prototype CEGAR tool. Haskell is suggested as the programming language in order to speed up development and to exploit local knowledge.","","","","If this topic is picked up by a group of 2 students then there is ample opportunity for benchmarking some novel abstractions that have only been postulated but not yet implemented. This could lead to a publication in the prestigious SPIN workshop proceedings."
"KE10","CocoaSpin","KE","Kai Engelhardt","","","Active","CECSSE","Development","","Spin is a popular model checker, ie a software tool for automatically verify that a system (given as a program text in the procedural language Promela which has constructs for a.o. message passing and shared variable concurrency) satisfies a property (given as a linear time temporal logic formula). Spin itself is a command line tool that runs n just about everything. Its most elaborate current interface is graphical yet arcane and written in Tcl/Tk. This is both ugly and unacceptable for Mac users. The aim of this thesis project is to radically redesign a GUI for Spin using a modern GUI libraries such as Cocoa on Mac OS X, incorporating graph layout algorithms from the graphviz project. (The Tcl/Tk interface xspin has an option to use graphviz to display the BŸchi automata related to the program text. CocoaSpin should be at least as good.)","","","",""
"KJE15","A Secure Bootstrapper for the seL4","KJE","Kevin Elphinstone","","","Active","CECSSE","R & D","","The seL4 microkernel is a high assurance microkernel capable of acting as a seperation kernel when it and the encompassing system is instantiated correctly. The goal of this thesis is to develop a simple component model that can specific an initial system state - i.e. the servers and applications that will run on the microkernel. THe component model is then used to generate the boot strapping code to instantiate the system with the specified seperation guarantees. The project may involve evaluating the existing CAMKES framework for the component model, and looking at formal models and guarantees for both the component model, and the generation of the boot strapper.","","","",""
"KJE16","Linux as a component","KJE","Kevin Elphinstone","","","Active","CECSSE","R & D","","NICTA has various versions of Linux that run para-virtualised on various versions of micro-kernels developed here at NICTA. However, the connection between Linux and the platform is rather ad-hoc, which makes is difficult bring Linux into the principled componet framework (CAMKES) developed here at NICTA. This project would involve examining the interface between the micro-kernel and the support infrastructure to allow Linux to be just another component in the CAMKES framework.","","","",""
"KJE17","ARTEMIS robotic clarinet player ","KJE","Kevin Elphinstone","","","Active","BINFBIOMCECSSE","R & D","","NICTA is entering the ARTEMIS intrument playing robot competition. This project involves developing the system software side of the robot, with an eye to making it general enough to use it for future entries. It involves low-level embedded controller programming, Linux kernel programming, and application programming. A familiarity with music is also helpful.","","","",""
"LZ01","Automated Web Service Composition using Semantic Web Technologies","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","One of the most important benefits of using Web services is to build new business processes, services and values through compositing existing services quickly and reliably. This requires not only syntax description of the web services but also semantic descriptions being integrated into the web services and used effectively in an automated composition process. You can choose one of the two focuses of this  project: 1) mapping process specific web service elements to semantic concepts 2) service composition using semantic concepts. The current research has been focusing on using ontology models for semantically annotating Web services. This project will expand the ontology model approach and directly leverage semantics in business process models. Model Driven Development (MDD) approach will be applied in the research to enable model mapping/transformation/merging between business process models, Web services models and ontology models.  This topic is closed related to emerging Web 3.0 technologies which enhances current  Web 2.0 technologies.  Students will be able to be exposed to next-generation AJAX and RESTful services. Students may work with one of Australia's leading e-Business standardization body to solve real-world problems as their case studies. The work will contribute to the standardization body directly. Students may also have the opportunity to do additional work, which will be qualified for ""Industry Training"" required by the school.Students will work closely with senior researchers at National ICT Australia in a very friendly mixed-gender team environment. Suitable for students interested in business process composition using web services. This topic will be co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). ","For more information: contact Liming Zhu limingz@cse.unsw.edu.au","","",""
"LZ02","Domain Specific Modelling Language Design for Model Driven Development","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","Domain Specific Modelling Languages (DSML) is about raising the level of abstraction for software development, providing more powerful domain specific concepts for capturing and reusing knowledge. It is the most critical part in adopting a Model Driven Development (MDD) approach (such as OMGÕs MDA or MicrosoftÕs DSL). Careful considerations need to be taken into account when a DSML is designed for a combination of different purposes, such as introducing new domain notation, domain specific analysis, eliminating repetitive tasks, facilitating data description or system configuration and making interaction programmable. There are trade-offs involved in fulfilling these purposes. This project is to investigate best practices in DSML design in different contexts and develop a framework and associated guidelines for choosing the right practices for maximizing the effectiveness of the final DSML. Two case studies from both development domains (e.g. web services modelling languages and pattern languages) and business domains (e.g. finance) will  be used to validate the framework.Students will work closely with senior researchers at National ICT Australia in a very friendly mixed-gender team environment. Suitable for students interested in software design and architecture and practical industry development methods. You will be exposed to cutting edge MDD practices in industry, in particular DSML design. This topic will be co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). ","For more information: contact Liming Zhu limingz@cse.unsw.edu.au","","","some useful links:http://www.codegeneration.net/http://msdn2.microsoft.com/en-us/teamsystem/aa718951.aspxhttp://www.omg.org/mda/"
"LZ05","Model Driven .Net Development","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","Model Driven Development (MDD) is about raising the level of abstraction for software development, providing more powerful concepts for capturing and reusing knowledge in platforms and languages. Different methodologies exist for practicing MDD, such as OMG's Model Driven Architecture (MDA) and Microsoft's Domain Specific Language (DSL). The inseparable link between models and code created through MDD demands consistent management of models and code in a single IDE. This research is to apply our existing experience in MDA to DSL-based MDD by developing domain specific designers as plug-ins for Microsoft Visual Studio. Students will use the DSL toolkit to design their own high level domain specific language (such as performance testing, pattern integration) and capture reusable assets and best practices in supporting infrastructure as libraries, code generation templates and guidance toolkits. Students will be exposed to leading edge research in software architecture domain and latest industry tools especially Visual Studio 2005 and its DSL/GAT toolkits. Students will work closely with senior researchers at National ICT Australia in a very friendly mixed-gender team environment. Suitable for students interested in software design and architecture and practical industry development methods. The research will improve the understanding of practical requirements for MDD practice in modern IDEs. It will also assist in capturing reusable assets and best practices as domain specific languages and supporting platforms. Students will have both theoretic and practical knowledge of the next generation model driven software development paradigm. Email: limingz@cse.unsw.edu.auThis topic will be co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). ","http://www.cse.unsw.edu.au/%7Elimingz/publication/WICSA2005_Zhu.pdf","","http://msdn.microsoft.com/vstudio/dsltools/ , http://msdn.microsoft.com/vstudio/dsltools/default.aspx?pull=/library/en-us/dnbda/html/softfact3.asp","some useful links:http://www.codegeneration.net/http://msdn2.microsoft.com/en-us/teamsystem/aa718951.aspxhttp://www.omg.org/mda/"
"LZ06","Model Driven Development for Web Service ","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","Model Driven Architecture (MDA) has proven useful in component technologies(.Net, J2EE, CCM). Service Oriented Architecture (SOA) is the new wave in enterprise application development and integration. But there are a few new challenges in SOA such as performance issues, lack of standards in development process and easily adopting existing systems to SOA. This research is to apply MDD to Web service development in terms of standardizing building and evolving service oriented system on the model level, conducting critical performance benchmarking before the full system is committed to implementation and adapting existing software systems (components, legacy systems) to SOA with minimum effort. Students will research SOA and MDA and develop model transformation utilities for SOA systems using existing open source frameworks. Students will be exposed to leading edge research in software architecture area and latest industry development paradigm and tools.  Students may work with one of Australia's leading e-Business standardization body to solve real-world problems as their case studies. The work will contribute to the standardization body directly. Students may also have the opportunity to do additional work, which will be qualified for ""Industry Training"" required by the school.Students will work closely with senior researchers at National ICT Australia in a very friendly mixed-gender team environment. Suitable for students interested in software design and architecture and practical industry development methods. The research will improve the understanding of practical requirements for MDA practice in SOA. It will also assist in capturing reusable assets and best practices as model transformation ""cartridges"". Students will have both theoretic and practical knowledge of the next generation model driven software development paradigm in the context of Web services. Email: Liming Zhu limingz@cse.unsw.edu.auThis topic will be co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). ","http://www.cse.unsw.edu.au/%7Elimingz/publication/WICSA2005_Zhu.pdf","","http://www.omg.org/mda/",""
"LZ07","Adaptive Software Process Engineering","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","Eclipse Process Framework (EPF) is an extensible framework for software process engineering. It provides facilities for process authoring, library management, configuring and publishing a process. The aim of the project is to evaluate EPF and provide further support on tracing of relationships between software process elements and other external elements. Students will be exposed to software process engineering and latest software development methodologies. This project will also be supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html)","contact limingz@cse.unsw.edu.au","","","http://www.eclipse.org/epf/"
"LZ08","Service Oriented Architecture for e-Business Standards","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","Recently, industry consortiums have been developing e-Business standards using  XML and business process modeling. Such standards inevitably have to be mapped on to technology layers such as service-oriented and event-driven infrastructures. Deriving a flexible reference architecture and implementation from e-Business standards are not always straightforward. It involves multiple technical and not-technical factors and design trade-offs. This project will guide students to look into the state-of-art in e-Business standards, web service protocol stacks and service coordination mechanisms.  Students will work with one of  Australia's leading e-Business standardization body to solve real-world problems by inventing methods and implementing prototypes. The work will contribute to the standardization body directly. Students may also have the opportunity to do additional work, which will be qualified for ""Industry Training"" required by the school. Most of students who are involved in this project previously have been offered a job in the financial industry.Students will work closely with researchers at National ICT Australia in a very friendly team environment. Suitable for students interested in software design, web services  and industry-scale development. For more information, please email: Liming Zhu limingz@cse.unsw.edu.au. This topic will be also co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html).","http://www.research.ibm.com/journal/sj/452/roth.htmlhttp://www.research.ibm.com/journal/sj/452/hinkelman.html","strong programming skills required.","","http://www.lixi.org.au"
"LZ09","Design-time Adaptive Service Engineering ","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","This project will investigate how to build adaptiveness  into a service oriented architecture to make the service more evolvable and flexible when facing changing business requirements. The project will guide student to look into the state-of-art in web service protocol stacks and design-time adaptation mechanisms.  Students may work with one of AustraliaÕs leading e-Business standardization body to solve real world problems by inventing methods and implementing prototypes. Students may also have the opportunity to do additional work, which will be qualified for ÒIndustry TrainingÓ required by the school. Students will work closely with researchers at National ICT Australia in a very friendly  team environment. Suitable for students interested in software design, web services  and industry-scale development. For more information, please email: Liming Zhu limingz@cse.unsw.edu.au. This topic will be also co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html).","","strong programming skills required.","","You can either choose to work in the .Net environment:http://msdn2.microsoft.com/en-us/teamsystem/aa718951.aspxOr other environment:http://www.codegeneration.net/"
"LZ10","Service-Oriented Eventing and Content Distribution","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","Web services often want to receive notifications and contents when events occur in other services. A mechanism for effectively registering specific interests and delivering tailored contents is needed.  While there are existing web service standard for this, the mechanism needs to be tailored and optimized to support domain specific event and content distribution.  This project will guide students to apply such a mechanism to a real world product information distribution scenario.   Students may work with one of AustraliaÕs leading e-Business standardization body. Students may also have the opportunity to do additional work, which will be qualified for ÒIndustry TrainingÓ required by the school. Students will work closely with researchers at National ICT Australia in a very friendly  team environment. Suitable for students interested in software design, web services  and industry-scale development. For more information, please email: Liming Zhu limingz@cse.unsw.edu.au. This topic will be also co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html).","","","http://www.w3.org/Submission/WS-Eventing/","students will be exposed to advanced standards in the web services domain. "
"LZ11","Architecture for Ultra-Large-Scale (ULS) Systems","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","R & D","","Vertical industries have been developing e-business standards to improve their business-to-business interoperability, and to foster efficient and organically-grown industry-wide systems. Such systems and their horizontal interactions with even larger ecosystems are effectively an Ultra-Large-Scale (ULS) system [1]. E-Business standards are essentially a set of governing rules for these ULS systems. It is critical that these governing rules address the socio-technical ecosystem (rather than system of systems) challenges recognized for ULS system, promote overall system quality, and cope with long-term, self-regulating, continuous evolution. In this project, students will investigate architectural issues unique to a ULS system. The role of architecture in this case is significantly different from traditional technical architecture which focuses on structural components and connectors. In order to address both the business perspective and ULS system challenges, the architecture needs to balance consistency and variety, address competing needs from different parties and consider trade-offs between prescriptive guidance and an ability to evolve. Students will work with one of Australia's leading e-Business standardization body to solve real-world problems. The work will contribute to the standardization body directly. Students may also have the opportunity to do additional work, which will be qualified for ""Industry Training"" required by the school. Students will work closely with researchers at National ICT Australia in a very friendly team environment. Suitable for students interested in software design, web services and industry-scale development. For more information, please email: Liming Zhu limingz@cse.unsw.edu.au. This topic will be also co-supervised by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html).","[1]  L. Northrop, R. Kazman, M. Klein, D. Schmidt, et al., ""Ultra-Large Scale Systems: The Software Challenge of the Future,"" 2006. http://www.sei.cmu.edu/uls/","","email limingz@cse.unsw.edu.au","http://www.sei.cmu.edu/uls/"
"LZ12"," Tools for Selection of Web Services using Business Value Information","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","Development","","Web services are distributed computing application components that use a number of Extensible Markup Language (XML) based technologies to implement the service-oriented architecture (SOA). Implementation-independence of Web services technologies allows different businesses to collaborate and achieve common business goals despite the fact that the collaborating Web services can be distributed over the Internet, run on different platforms, and implemented in different programming languages. Web services technologies are already embedded in various products and services of all major computing companies and used for diverse purposes.   The Universal Description, Discovery and Integration (UDDI, http://www.uddi.org/) is an industrial standard for describing businesses and their Web services and allows discovery and selection of Web services during run-time. This supports flexible and adaptive run-time composition of Web services to address changing business needs. However, UDDI describes only ÒwhatÓ Web services do and not Òhow wellÓ they do it. As the number of Web services that offer similar functionality increases in the global market, comprehensive description of their quality of service (QoS, such as response time, availability) and various metrics of business value (e.g., price, price/performance ratio, customer satisfaction) will become major competitive advantages. In many cases, Web services will provide the same or very similar operations, but with different QoS and business value. While several UDDI extensions specifying QoS information appeared recently, specification of business value is still relatively superficial.   Students working on this thesis project will design, implement, and test an existing open-source UDDI implementation with capabilities for publication and search of various business value measures (and, possibly, other relevant information, such as QoS). They will also design, implement, and test tools with novel algorithms for selection among similar Web services to maximize various business value metrics. ","http://www.uddi.org/","","http://searchwebservices.techtarget.com/originalContent/0,289142,sid26_gci916789,00.html","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ."
"LZ13","Extending Eclipse for Specification of Business Value Information","LZ","Liming Zhu","","","Active","BINFBIOMCECSSE","Development","","   The Unified Modeling Language (UML) became a widely used industry standard for specification of design-time software models. On the other hand, the Business Process Modeling Notation (BPMN) is an industry standard for specification of business process models. Its popularity rapidly increases. Unfortunately, neither UML nor BPMN does not enable specification of very important design information related to expected business value (e.g., price, price/performance ratio, customer satisfaction) and desired quality of service (QoS, such as response time, availability).   Students working on this thesis project will work with Eclipse IDE (Integrated Development Environment) support for UML support for either UML or BPMN. They will design, implement, and test new plug-ins for the Eclipse IDE (Integrated Development Environment) to enable specification of business value measures (and, possibly, other relevant information, such as QoS) in UML or BPMN models. If time allows, they might also design and implement novel algorithms for exporting/importing the resulting annotated UML or BPEL models and for processing of these exported files. This is a development thesis project, where the focus is on implementation and hands-on experiments, instead of surveys and analysis of research literature. ","http://www.eclipse.org/","","","The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ."
"MEB10","Modelling Musical Improvisation","MEB","Mike Bain","","","Active","BINFBIOMCECSSE","Research","","Modelling musical knowledge to the level of a skilled human is a challenge forartificial intelligence.  Many approaches have been investigated towards thisend, e.g., for modelling composition, stylistic expression and performanceskill.  Representations such as grammars and techniques such as neural networksand genetic algorithms have been used.A particularly well-studied approach to modelling musical performance hasadopted a rule-based approach.  Existing systems can be given a written scoreand produce a musically-acceptable performance.  Compositional systems that usesets of rules to produce pieces ""in the style of"" noted composers have alsobeen developed.In this project the goal is to construct a system for jazz improvisation, anactivity which lies somewhere between the tasks of performance and composition.The hypothesis is that successful improvisation over a set of chord changes canbe achieved by selecting from a given set of note patterns or ""lines"" andcombining them using suitable ""conjunctions"" to form a musically-acceptableperformance.The project will involve a number of stages: research into musicalrepresentation by rules; eliciting a set of lines from a domain expert;constructing a Markov chain model for linking lines to be played over a givenchord sequence; innovating a novel representation and sequencing technique for""smoothing"" the connection between lines by adding conjunctions.Applicants must have good programming skills (e.g., in Java).Fluency in reading musical notation is an advantage.The project will be jointly supervised by CSE and the School ofMusic and Music Education.","","","","Applicants must have good programming skills (e.g., in Java).Fluency in reading musical notation is an advantage.The project will be jointly supervised by CSE and the School ofMusic and Music Education."
"MH01","Automatic trip diary for mobile phones","MH","Mahbub Hassan","","","Active","CECSSE","R & D","","The student will build an application for mobile phones that will keep track of the locations visited by the phone and display the past trips when requested. The prototype will be built on a N95 NOKIA GPS mobile phone running Symbian OS.","","","",""
"MH02","Drive-by Mobile Imaging System","MH","Mahbub Hassan","","","Active","CECSSE","R & D","","The student will develop a mobile imaging system that can automatically capture, tag, store, search and retrieve images from a dash-cam in a moving vehicle. The prototype will be built on a N95 NOKIA GPS mobile phone with in-built 5 megapixel camera and running Symbian OS.     ","","","","The student will need valid driving license for data collection and testing of the prototype."
"MIJM02","Scheduling using Constraint Programming","MIJM","Michael Maher","","","Active","All Students","R & D","","Scheduling problems require us to allocate to different tasks a time to use different resources.  In industrial optimization the resources are machines and the tasks are manufacturing jobs.  For example, a steel mill will have heating processes, milling machines, etc., which are used in different combinations to manufacture different products.  In calendar applications the resource is the attention of a person and the tasks are meetings.  In exam scheduling, resources are rooms and tasks are examinations.The aim of scheduling is to make best use of the resources so that as many tasks as possible are completed in time.  Finding an optimal schedule is NP-hard, in general.  Nevertheless, constraint programming techniques have been successful in solving many practical scheduling problems.The main focus of this project is to investigate and evaluate constraint programming, local search, and hybrid techniques for specific kinds of scheduling problems.  Students will have the opportunity to work on scheduling problems arising from one of NICTA's industrial partners.","","","",""
"MIJM03","Using constraints to define permitted traffic flows","MIJM","Michael Maher","","","Active","All Students","R & D","","NICTA and industrial partner are defining permitted road traffic flows using constraints. This project involves the development of a tool that will generate the appropriate constraints for an intersection based on inputs from a traffic engineer. The project involves (a) requirements gathering from the industrial partner to determine what inputs are needed and how they determine the generated constraints, (b) design and implementation of a user interface to accepts these inputs, and (c) generation of the constraints.","","","",""
"MM04","3D immersive web-browsing","MM","Matthew McGinity","","","Active","BINFBIOMCECSSE","R & D","","This project will tackle the problem of embedding dynamic web content in a 3D virtual reality environment.  The VR theatre is iCinema's A.V.I.E - a 360¡ cylindrical stereo projection system (www.icinema.unsw.edu.au).  The project will look at the technical and design issues involved in navigating and interacting with web content in such an environment.","","Strong C++ and computer graphics skills absolutely necessary. Microsoft Visual Studio and some Windows programming.","",""
"MM05","3D GUI for virtual reality environments","MM","Matthew McGinity","","","Active","BINFBIOMCECSSE","R & D","","This project investigate the design and development of 3D graphical user interfaces for use in iCinema's A.V.I.E - a 360¡ cylindrical stereo VR theatre (www.icinema.unsw.edu.au).  The project will consider usability, technical and aesthetic issues involved in the design of 3D interfaces for a variety of different tasks, using the theatre's full-body camera-based tracking system and pointing devices as interface devices. The project should also consider the problem of multi-user interaction.","","Strong C++ and computer graphics skills necessary. Microsoft Visual Studio and some Windows programming required.","",""
"MM06","Navigating virtual earth databases in a VR environment","MM","Matthew McGinity","","","Active","BINFBIOMCECSSE","Development","","This project will develop a means of embedding and navigating the databases of Google Earth or Microsoft Virtual Earth in a 360¡ panoramic VR theatre (see www.icinema.unsw.edu.au).  The project will investigate the engineering issues involved in streaming and rendering the 2D, 3D and panoramic data available in these databases, as well as the human-computer interface design issues in navigating such a database in this environment.","","Strong C++ and computer graphics skills necessary. Microsoft Visual Studio and some Windows programming required.","",""
"MMTC18","3D Computer Games and Functional Programing","MMTC","Manuel Chakravarty","","","Active","CECSSE","R & D","","The goal of this thesis is to investigate the advantages and disadvantages of modern functional programming in the development of 3D computer games. Functional programming languages have demonstrated a gain in productivity in a range of application areas (e.g., Ericsson's use of the language Erlang for telecommunication). Todays computer games are very complex pieces of software and any advance in language technology that simplifies the development of computer games would be a welcome help. Moreover, improvements in this area promise to be of economic relevance as the entertainment industry, in general, and the computer games industry, in particular, are profitable industries with a health growth rate.This is a continuing topic, where students add functionality to the 3D first-person shooter ""Frag"" implemented in Haskell.  So far, Frag went through two iterations.  The first build the basic game from scratch including support for standard file formats needed for a 3D game and a novel approach to entity control.  The second restructured the game and added NPC AI.","","","",""
"MMTC20","Bridge between Objective-C and Haskell: The next generation","MMTC","Manuel Chakravarty","","","Active","CECSSE","R & D","","The aim of this project is to develop an interoperability bridge between two rather different languages, namely Objective-C 2.0 and the purely functional language Haskell.Functional languages, such as Haskell, have recently received growing industrial interest.  This is due to improved productivity and easier reasoning about the behaviour of programs.  Moreover, the recent focus on parallel programming, trigged by multi-core and many-core processors, boosted the visibility of functional languages due to their controlled use of side effects.Objective-C 2.0 is the latest version of the language and the main development language supported by Apple for Mac OS X.  Hence, access to Objective-C 2.0 libraries from Haskell is very valuable for software development with Haskell for Mac OS.Challenges include arranging for cooperation between the separate storage management systems included in the runtimes of Haskell and Objective-C 2.0 as well as integrating the type and overloading system of Objective-C in Haskell for convenient library access.","","","","Old bridge: http://hoc.sourceforge.net/Previous experience with Objective-C would be a bonus.  This topic requires existing access to a Mac running Leopard."
"MMTC21","High-performance computing on your graphics card: physical simulations on modern GPUs","MMTC","Manuel Chakravarty","","","Active","CECSSE","R & D","","GPGPU stands for General-Purpose computation on GPUs. With the increasing programmability of commodity graphics processing units (GPUs), these chips are capable of performing more than the specific graphics computations for which they were designed. They are now capable coprocessors, and their high speed makes them useful for a variety of applications.  High-end GPUs deliver a higher peak performance than high-end CPUs at a lower price point.  The problem is that they are more difficult to program and require very highly data-parallel algorithms.In this project, we focus on the use of GPUs for physical simulations. Applications range from physics engines in modern 3D games to computational science.  The implementation will be based on CUDA (""Compute Unified Device Architecture"").  CUDA is a GPGPU technology that allows a programmer to use the C programming language to code algorithms for execution on the GPU. CUDA has been developed by Nvidia and to use this architecture requires an Nvidia GPU and special stream processing drivers.","","","","Further reading:* http://gpgpu.org/* http://www.nvidia.com/object/cuda_home.html"
"MMTC22","High-performance computational chemistry with Monte-Carlo simulators","MMTC","Manuel Chakravarty","","","Active","BINFBIOMCECSSE","R & D","","The aim of this project is to further develop the worlds fastest Monte-Carlo simulator for polymer kinetics (i.e., the formation of polymers from monomers).  The simulator, itself written in Haskell, uses a generative approach by compiling a high-level description of a chemical model into highly optimised low-level C code.  This work address the tension between software generality and performance in the  domain of simulations based on Monte-Carlo methods.  It simultaneously  achieve generality and high performance by a novel development methodology  and software architecture centred around the concept of a specialising simulator generator.  The approach combines and extends methods from functional programming, generative programming, partial  evaluation, and runtime code generation.  We also generate parallelised simulators.The  simulator so far is for advanced forms of polymerisation kinetics.  It achieves unprecedented performance, making Monte-Carlo methods practically useful in an area that was previously dominated by deterministic PDE solvers.  This is of high practical relevance, as Monte-Carlo simulations can provide detailed microscopic information that cannot be obtained with deterministic solvers.","","","","Background in Chemistry would be an advantage, but is not strictly necessary.Further background: http://www.cse.unsw.edu.au/~chak/papers/KCCSB07.html"
"MMTC23","Haskell support for Xcode","MMTC","Manuel Chakravarty","","","Active","CECSSE","Development","","Xcode is Apple's IDE included with every Mac.  It supports a range of languages out of the box and includes a plugin interface to add support for other languages.  It is the aim of this project to add support for the functional language Haskell.The exists a incomplete Xcode plugin for Haskell developed previously outside of UNSW, which may be used as the basis for this work.  A goal of this project would be to use the external API of the Glasgow Haskell Compiler to realise the plugin.  This API enables extracting detailed information out of a given Haskell program, including full type information, even for code where type inference has to be performed to gain this type information.Functional languages, such as Haskell, have recently received growing industrial interest.  This is due to improved productivity and easier reasoning about the behaviour of programs.  Moreover, the recent focus on parallel programming, trigged by multi-core and many-core processors, boosted the visibility of functional languages due to their controlled use of side effects.","","","","Further information: http://maxao.free.fr/xcode-plugin-interface/ and http://haskell.org/"
"MP07","Coordinating teams of robots","MP","Maurice Pagnucco","","","Active","All Students","R & D","","This project deals with the coordination of two or more mobile robots in completing a complex task. In some cases the robots may have different abilities and are not capable of completing the task alone but can do so by coordinating with another robot. In other cases, coordination will speed the completion of the task.The first part of this project will involve developing a language that will enable the robots to communicate with one another. The second part will use this language for coordination and negotiation applied to a package delivery task.","","","",""
"MP11","Using negotiation to coordinate agents","MP","Maurice Pagnucco","","","Active","All Students","R & D","","Coordination of multiple entities is an important task for achieving optimal behaviour in a system. This is evident in many areas such as team games (e.g., basketball, soccer, etc) but also others such as courier services, taxi services, etc.The aim of this project is to develop a protocol for multiple agents that allows them to communicate and negotiate so as to coordinate their activities. The main tasks are: 1) develop or adapt an existing agent communication protocol; 2) develop a way for the agents to negotiate using this protocol, and; 3) apply this framework to a particular task such as coordinating multiple taxis to demonstrate how it works and assess its effectiveness.","","","",""
"MP17","Controlling a mobile robot using speech","MP","Maurice Pagnucco","","","Active","All Students","R & D","","This project aims to develop a speech control system for an ActivMedia PeopleBot mobile robot. An open source speech engine will be used to register spoken commands to the robot. These commands will be used to make the robot perform a series of predetermined tasks.","","","",""
"MP19","Using Speech to Play Games","MP","Maurice Pagnucco","","","Active","All Students","R & D","","Research in speech recognition has led to software capable of recognising human speech to a fairly high degree of accuracy. This includes open source software like Sphinx (http://cmusphinx.sourceforge.net/html/cmusphinx.php) and commercial software like Dragon Naturally Speaking. Speech often provides a more natural human-computer interface for human users. This project aims to provide a speech interface to the Half-Life game engine (http://www.valvesoftware.com/) that will be used to control virtual characters.","","","",""
"MP20","Cognitive vision","MP","Maurice Pagnucco","","","Active","BINFBIOMCECSSE","R & D","","This project aims to extract logical descriptions from camera images so as to reason about the objects that are in view. This information can then be used to complete goals. For example, in directing a robot to manipulate objects within its environment.","","Artificial intelligence with some logic.","",""
"MP21","Automatic Control of a 4-bladed Model Helicopter","MP","Maurice Pagnucco","","","Inactive","All Students","R & D","","This project looks at the automated control of a Draganflyer IV flying helicopter (see resources) from a computer using a PC and radio-control hardware (the PCBuddy). <p>Previous projects have replaced the standard flight control unit with a programmable unit equipped with gyroscopes and an Atmel AVR processor. <p>What this project will focus on is the development of software, through behavioural cloning for actually flying a path set by the user, or alternatively, following a ground target. <p>Vision information would be transferred back to the PC via a wireless video link. Subsequently and if time permits, other behaviour may be added, e.g. identifying objects on the ground. This could then be used for challenges such as Robocup Rescue Leage to identify victims in disaster situations. ","http://www.rctoys.com/draganflyer4.php<p>http://www.rc-electronics.co.uk/pcbuddy.htm<p>http://waleed.web.cse.unsw.edu.au/new/rescue.htmlhttp://robotarenas.nist.gov/competitions.htm","Recommended: COMP4411 experimental robotics. Knowledge of hardware required. DN average preferred. ","","Could be a one person or two person project. Students are encouraged to discuss with me if they have their own ideas for theses. WAS MWK01"
"MP22","RoboCup Standard League","MP","Maurice Pagnucco","","","Active","BINFBIOMCECSSE","R & D","","This topic is for those on CSE's RoboCup team competing in the Standard League competition using the humanoid Nao robot.Please contact me before nominating this topic.","","","",""
"MT01","Bacterial genomes: modelling the evolution of gene clusters","MT","Mark Tanaka","RL","Ruiting Lan","Active","BINFCS","Research","","A curious property of bacterial genomes is that genes involved in acommon pathway are often located near each other on thechromosome. Because such genes are not necessarily co-transcribed, itis not immediately obvious why this clustering shouldoccur. Alternative hypotheses have been advanced for the evolutionarymaintenance of bacterial genome organisation. For example, one theorystates that proximity of genes gives advantage to the cluster itselfthrough its ability to move between genomes via horizontaltransfer. Another theory claims that the ability for clustered genesto undergo amplification together confers advantage to thegenome. This project will involve using mathematical and computersimulation models to examine and distinguish between these alternativehypotheses.","Lawrence, JG, 2003, Gene organization: selection, selfishness, andserendipity. Annu Rev Microbiol 57:419-440.","","",""
"MT02","Evolution of Hepatitis C Virus and the role of escape mutants","MT","Mark Tanaka","","","Active","BINFBIOM","Research","","Around 170 million individuals worldwide are infected with HCV,including more than 200,000 Australians. Current treatments are not veryeffective and vaccines are not yet available. Due to its rapid mutationrate, HCV is able to evolve fast, avoiding recognition by the hostimmune response. Escape mutants evading recognition can potentiallyspread in the population and therefore represent a major obstacle to thedevelopment of effective vaccines. Identifying the factors that shapegenetic diversity in rapidly mutating viruses remains a formidablechallenge. This project will assess the following goals through abioinformatics approach.1) To apply available algorithms to predict the distribution of HCV epitopes (parts of viral proteins recognised by the immune system) that are presented on the MHC class I molecules. These predictions will be used to test whether locations of predicted epitopes are correlated with the variability of the virus.2) To investigate the effect of host genetic factors (MHC frequencydistribution within populations) in shaping the diversity ofHCV. Here, we quantify the genetic diversity of HCV strainscirculating among different human populations, using HCV databases.","","","",""
"MT03","Evolution of viruses within hosts","MT","Mark Tanaka","RL","Ruiting Lan","Active","BINFBIOM","Research","","Viruses replicate fast, exploiting cellular resources and processes asthey invade host organisms.  Constraints in this process at themolecular level may give rise to evolutionary trade-offs. A currentHonours student in our group is investigating the evolutionaryoptimisation of viral replication within cells. That project can beextended to consider the dynamics of viral growth between cells withinhosts. What host factors - particularly those relating to the immunesystem - influence the replication ""strategies"" of viruses?  How dothese factors affect the rate of viral sequence evolution?  Thisproject will involve mathematical and/or computer simulation modellingto study viral evolution within hosts. It may also involve theanalysis of DNA or protein sequences.","Regoes, RR, Antia R, Crotty S and Tanaka MM, 2005, Optimal replicationof poliovirus within cells, Am Nat 165:364-373.","","",""
"MT04","Simulating the extremes in combinatorial libraries and repertoires.","MT","Mark Tanaka","","","Active","BINFBIOM","Research","","This project aims to understand the properties of extreme phenotypesin large random libraries or repertoires of combinatorially producedmolecules. There are two important applications: first, aptamers,which are biotechnological constructs and second, B cells, which arecomponents of the immune system producing antibodies.  There areparallels between these two disparate systems that share the commonproperty of relying on combinations of units to explore phenotypespace.  How effective (in terms of affinity to a target) is the bestspecies in a library/repertoire? How quickly does the maximum rise asthe library/repertoire size increases?  Aptamers are developed througha process of selective enrichment. In the case of the immune system,antibodies increase in effectiveness through a process of sequentialimprovement in lymphoid tissues. This project will explore theseprocesses through computer simulations.","","","",""
"MTZ01","YOU STINKA","MTZ","Maria Tzortzis","","","Inactive","BINFBIOMCESE","Research","","  (eg AN01)* Topic Title	Supervisor Initials	MTZAssessorSession/s Topic is Available	 s1 only  s2 only  Either Session  Unique  Not Available  Archive Note: Archive will effectively delete topic from list.* Topic is Available To	 All Students  BINF Priority Note: only topics with a biological context should be entered as BINF Priority. See Bruno Gaeta for more info.* Topic is of Type	 Research  Development  R&D  COR Enter COR Code:Is this topic an Individial Thesis or a Group Thesis?	 Individual  Group Is this topic an Industrial Topic	 Yes  No Show topic on web?	 Yes  No DescriptionMake sure the description is over 150 chars or else wont be shown on web.","","","",""
"NHL08","Visual Cortex Image Processing System for Retinal Prosthesis","NHL","Nigel Lovell","","","Active","BINFBIOMCECSSE","R & D","","A software system to perform image acquisition, processing and rendering for a vision prosthesis (bionic eye) is under development. While some aspects of the system have been completed, little work has been done on image processing and rendering algorithms. This will be the focus of this project.","","","",""
"NM03","When and how to use animation: Connecting human minds to computers.","NM","Nadine Marcus","","","Active","All Students","Research","","This project will involve creating and comparing two simple websites/applications/tasks with and without animation.  Some user data will then need to be collected and statistically analysed to establish conditions under which animation is most useful.","Sweller, J.  1999. ""Instructional Design in Technical Areas"".  ACER Press: Camberwell, Victoria.","Completion of a Human Computer Interaction or Psychology course is essential.  An ability to conduct a basic scientific study and a willingness to do some background reading into Cognitive Load Theory is needed.  Skills in producing basic graphics/animation will also be needed.  ","","This topic will be co-supervised by Dr Paul Ayres an  expert in instructional animations within the School of Education Studies."
"NM05","Designing Adaptive Interfaces","NM","Nadine Marcus","","","Active","All Students","R & D","","The project will involve creating at least 2 different interfaces for some basic computing gaming software.  One interface will be tailored to more novice users and the other for experts.  User testing will then need to be conducted using a group of both experts and novices.  Statistical analyses will need to be carried out as well.","Kalyuga, S., Ayres, P., Chandler, P., & Sweller, J. (2003).  Expertise reversal effect.  Educational Psychologist, 38, 23-31.Kalyuga, S. & Sweller, J. (2005) Rapid Dynamic assessment of expertise to improve the efficiency of adaptive e-learning.  Educational Technology Research and Development, vol 53, p83-93.","Human Computer Interaction or Psychology course.","","A basic understanding of how to conduct empirical research is essential.  Some understanding of psychology research principles will be needed."
"NM06","Adaptive E-learning","NM","Nadine Marcus","","","Active","All Students","R & D","","The project involves using AI-based planning and machine learning techniques to build flexiblee-learning systems.  A system will be developed that can adapt to the learners level of knowledge in a domain.  ","","A WAM of distinction level or higher.","","This topic is jointly supervised with Claude Sammut"
"NM09","Next generation interface design and development in an adaptive eLearning environment. ","NM","Nadine Marcus","","","Active","All Students","R & D","","This honours project is a part of an exciting initiative that involves R&D of an Intelligent Tutoring System that is capable of adaptation of delivered content to user's level of knowledge. We are exploring rich visual modes of interaction between users and our intelligent eLearning environment.Your focus will be on designing and creating next generation interfaces (think Mac OSX and MS Office 2007), which you will then need to evaluate with real users.This is an excellent opportunity for someone who is interested in real world development and design experience and taking part in a research that spans fields such as Human Computer Interaction, Artificial Intelligence and Cognitive Psychology.","","Background in Visual or Graphics Design, Human Computer Interaction or Psychology.Experience in web design using Adobe Flash would be valuable. Some knowledge of dynamic interpreted  programming languages such as ActionScript, JavaScript, PHP, would be also be useful. Need a WAM of 70 or higher.","","This honours project will allow you to take part in start of the art research that is currently being funded by the engineering faculty. "
"PB01","Defining Software Project Success","PB","Paul Bannerman","","","Inactive","CECSSE","Research","","There are many criteria that are used to determine whether a project is a success or not.  Each approach has benefits and limitations.  There is no consensus in the literature on the best approach.  This project offers the opportunities to investigate these options and propose an approach that you can justify as being most suitable for software projects.","","","","This topic is the same as DRJ08. Dr Paul Bannerman is a NICTA research scientist and conjoint UNSW/CSE research fellow. You will need to attend review meetings at the NICTA ATP laboratory near the Redfern train station."
"PB02","Discontinuities in software development","PB","Paul Bannerman","","","Active","SE","Research","","This is a research topic that blends theoretical and practical analysis. The aim is to identify disruptions (discontintuities) that can occur during software development projects and consider the impact(s) they might have on outcomes. The analysis might investigate individual discontinuities or different categories of discontinuities.","","","","Dr Paul Bannerman is a NICTA research scientist and conjoint UNSW/CSE research fellow. You will need to attend review meetings at the NICTA ATP laboratory near the Redfern train station."
"PB03","Determinants of project success","PB","Paul Bannerman","","","Active","CECSSE","Research","","This is a research topic on the differential determinants of project success. The literature talks about success factors, processes, and capabilities as being key to software project success. How can these alternative explanations be rationalised? (Will the real determinants of project success please stand up!)","","","","Dr Paul Bannerman is a NICTA research scientist and conjoint UNSW/CSE research fellow. You will need to attend review meetings at the NICTA ATP laboratory near the Redfern train station."
"PB04","Risk management in projects of different type","PB","Paul Bannerman","","","Active","CECSSE","Research","","This is a research topic on risk management in software projects. Recent research has identified that not all software projects conform to a generic type assumed by project management prescriptions (such as PMI's PMBOK Guide). For example, one study identified four types: 'pure' project from; hybrid form, operational activity; and breakthrough event. The aim of the project is to investigate how risk management strategies and practices might vary across software projects of different types.","","","","Dr Paul Bannerman is a NICTA research scientist and conjoint UNSW/CSE research fellow. You will need to attend review meetings at the NICTA ATP laboratory near the Redfern train station."
"PB05","Methodologies: When are they useful?","PB","Paul Bannerman","","","Active","CECSSE","Research","","This is a research topic that aims to assess the value of methodologies in software engineering (software development, project management, risk management, etc.). Are methodologies beneficial? If so, why and when? What are their limitations and how might they be overcome?","","","","Dr Paul Bannerman is a NICTA research scientist and conjoint UNSW/CSE research fellow. You will need to attend review meetings at the NICTA ATP laboratory near the Redfern train station."
"PB06","Knowledge transfer in software development teams","PB","Paul Bannerman","","","Active","SE","Research","","What are the mechanisms whereby knowledge or expertise are transferred within software development teams? What factors influence knowledge transfer within software development projects? For example, what influence does the software development methodology have on knowledge transfer?","","","","Dr Paul Bannerman is a NICTA research scientist and conjoint UNSW/CSE research fellow. You will need to attend review meetings at the NICTA ATP laboratory near the Redfern train station."
"RJB39","Dr Worm - using mobile phones in lectures","RJB","Richard Buckland","","","Active","BINFBIOMCECSSE","R & D","","Most students have mobile phones.  This project is to design and construct a system to allow these phones to be used to give instantaneous feedback in lectures, meetings etc etc  For example they could drive a ""worm"" like the one channel 9 used to assess the leaders debate of the 2007 federal election.","","3rd year networking or equivalent","","You'll need to be smart and motivated.  This could likely be turned into a commercial product.  It would probably be best for a group of 2 or 3 students, but if you have an alternative proposal feel free to pitch it to me."
"RJB40","A decent spell checker for dyslexics and bad spellers","RJB","Richard Buckland","","","Active","BINFBIOMCECSSE","R & D","","There is a reasonable amount of research on how dyslexics and poor spellers go wrong but still most spell checkers are not very good at helping them.  Two promising approaches are a personalised spell checker which learns over time, and using a phonetic dictionary as well as a traditional word list dictionary.This project is to investigate the current research and design a more effective spell checker for such people.  You'll need to have good coding and data structure and algorithm skills.  If you want to do the project as a team (my preference) then the aim is to produce an actual  usable spell-check plug in for some common linux or windows program or ajax app.  If you want to do the project as an individual then you should identify the part of the problem which interests you the most.","","","","You'll need to be smart, enterprising, and creative.  If you are interested in doing so we'll seek live spelling and misspelling data from google and perhaps interest them in the spell checker itself."
"RJB41","The CSE Computer Forensics Lab","RJB","Richard Buckland","","","Active","BINFBIOMCECSSE","R & D","","Design, cost, and construct a Computer Forensics Lab.This is a fairly unique project.  The computing security research group wants to create a world class digital forensics lab to teach our students forensics techniques and as a base for forensics research.  We have ample funding.  We need a carefully designed and budgeted lab.  The successful team will plan the lab, select the techniques and features most important to include, identify research and select appropriate hardware and software.  If the design is ready by the end of the first session then the team will also be able to construct and commission the lab.  The design will need to consider ethical and safety issues as well.","","","","This should be an extremely fun and satisfying project which will involve learning a great deal about computer forensics and current practise as well as overseeeing a real-world project.  Teams which wish to apply for the project need to produce a 10 minute design pitch outlining why they should be awarded the project.  This will be assessed in the same way as if your group was applying for a commercial tender.  Groups of size one will also be considered."
"RJB42","Hardware Virtualisation Rootkits","RJB","Richard Buckland","","","Active","BINFBIOMCECSSE","R & D","","Using hardware support on the current crop of intel and amd chips there is now the possibility of truly undetectable rootkits.  Investigate and explore the ways of detecting and preventing rootkits which use these features.  ","","","","Honours student Fionnbharr Davies has alrady developed a very nice proof of concept rootkit ""The Flying Spaghetti Hypervisor"".  Your mission, should you choose to accept it, is to examine this rootkit for weaknesses and work out ways to detect/thwart it and the techniques it uses.  This is cutting edge research!  You will need to have a high WAM.  Suitable for group or solo."
"RJB43","Network Forensic Analysis","RJB","Richard Buckland","","","Active","BINFBIOMCECSSE","R & D","","One of the major financial services organisations in Australia will provided access to all their network traffic and records.  You will forensically analyse this for possible intrusions or suspicious behaviour.  You may also design construct and deploy honeypots or other network sensors on their network to gather additional data.  Report back to the company on what you find and make recommendations as you see fit.","","A high WAM.  A strong background in networking, and COMP3441 Computer Security and Cryptography or equivalent. ","","Fancy being a sleuth? If you are interested in working in computer security or network forensics then this is the project for you.  A well done thesis will most likely open the door to any number of great jobs.  Group or solo."
"RJB44","Remote covert Jukebox ","RJB","Richard Buckland","","","Active","BINFBIOMCECSSE","R & D","","A partner in a high-powered city law firm has a Jukebox in his office.  The other partners aren't so pleased about it but he loves it.  I have talked him into funding a thesis project on his jukebox.  Your mission will be to internet-enable his jukebox in some useful and reliable and funky way.  Perhaps to be viewed in real time, perhaps to be heard and perhaps programmed remotely via a web page and/or some other ip based method which he could access from his blackberry or pda while he is away from his office. The task will involve meeting with the (very nice) guy, working out the requirements, prototyping and then building the system.  He'll purchase any hardware you require.  This is a very open ended and amusing project mimicing the product design process for a startup.  It would suit an entrepreneurial student / team with strengths in networking and web design.  As the system will have to operate stealthily though the company firewall there is potentially a security component to the design.  ","","","","picture here: http://www.cse.unsw.edu.au/~richardb/images/jukebox.jpg"
"RJB45","Computing for conservation and land management in the Megalong Valley.","RJB","Richard Buckland","","","Active","BINFBIOMCECSSE","R & D","","It constantly amazes me how many ways computing technology could be harnessed to assist and improve land conservation and bush regeneration, yet no one has put two and two together and invented and implemented them yet.For example there is a pressing need for cheap low powered remote and unattended monitoring and control systems, image processing and analysis, data logging and analysis, environmental mapping and interlinks with google mapping, low powered microcontrollers for optimum energy transmission from solar cells, and myriad other computing related products which haven't yet occurred to anyone.This task is open ended.  We have a (beautiful) large remote site in rugged semi-wilderness which is to be regenerated.   You will investigate the site and the problems posed and design some technological product to help in the conservation.Ideally your team will develop something simple and cheap which can then be used more widely by conservationists and land managers across the country.You will interview all the stakeholders, and then brainstorm a design for a product you could produce to help in managing and monitoring environmental regeneration sites.  This is an entropenurial project - you have to spot a niche need, think of an effective product, and then construct it.  You would then construct and test your prototype in the Megalong Valley.  The site is comprised of mix of old low intensity farmland and untouched woodlands in fairly rugged mountainous country near the coxes river.  As this is one of the major feeds for Warragamba Dam another option (which the Sydney Catchment Authority would love someone to pursue) is to look at remote unattended water quality and erosion monitoring - perhaps using wireless sensor networks. If you are an entrepreneurial individual team with technical skills and a desire to make a difference then come and talk to me and we can work out a customised thesis topic in this project which fits your skills and interests.","","","",""
"RL07","Bioinformatic analysis of pathway and regulatory networks of Shigella","RL","Ruiting Lan","","","Active","BINFBIOM","Research","","Shigella is an enteric bacterial pathogen that causes shigellosis. Shigella has been classified as a genus with 4 species. Our studies have shown that Shigella strains have been evolved from an E. coli ancestor at least 7 times and therefore are in fact pathogenic E. coli. However Shigella strains are known to be biochemically inactive, that is, they are unable to utilize many sugars as food source. The loss of these abilities must have occurred independently in different lineages.  Four shigella genomes from different lineages have now been sequenced. Through comparison of the four Shigella genomes and other E. coli genomes one can determine what metabolic pathways and regulatory networks that have been disrupted. A previous bioinformatics honours student has developed an approach to collate pseudogenes data. This project will build on that study and further develop methods and programs to collate and analyse pathway and regulatory network data.","","","",""
"RL08","Design and construction of a genotyping database for evolutionary and epidemiological tracing of bacterial pathogens","RL","Ruiting Lan","","","Active","BINFBIOMCS","R & D","","Our laboratory is currently developing genotyping methods for a number of bacterial pathogens, in particular for whooping cough causing bacterium Bordetella pertussis. We are generating genotyping data from a large number of isolates and aiming to use the data as a basis to build a database for global epidemiology and evolutionary analysis of B. pertussis. The bioinformatics honours student will join the research team to design and construct the genotyping database and develop tools to analyse the genotype data.","","","",""
"RL09","Development of an integrated environment for sequence editing and comparison","RL","Ruiting Lan","","","Active","BINFBIOMCS","R & D","","Repeat sequencing of many samples to reveal variation at sequence level is used for population, evolutionary and ecological studies and many other applications. The whole process from editing of raw sequence data to sequence comparison often uses different programs. We are currently using third party programs for editing and a program we developed for sequence comparison. This project is to design and implement an integrated environment for this process.","","","",""
"SK001","Scalable Network Layer Multicast for Wireless Sensor/Actuator Networks","SK0","","","","Active","All Students","R & D","","Wireless Sensor Networks have been touted as ""one of the twenty emerging technologies that will change the world"" by MIT review, and ""one of the eight technologies to save the world"" by CNN. Though the technology is still in its early days, the range of potential applications is mind-boggling --- track bushfires, microclimates and pests in vine-yards, monitor the nesting habits of rare sea-birds, and control heating and ventilation systems, let businesses monitor and control their work spaces etc. Our overall goal is to develop new algorithms to optimize the performance and power consumption of highly distributed networked sensor devices. Undergraduates will help program at both wide-usage network simulation software (NS-2) developed at UC Berkeley and special, extremely tiny sensor devices, the wireless Telos MOTES (available from Moteiv Inc). This project is an excellent opportunity for students to gain both academic and industrial experiences working on a cutting-edge research topic jointly supervised by staffs from CSE and CSIRO ICT centre. The aim of this project is to design and evaluate a scalable network layer multicast protocol for sensor networks by simulations or/and experiments. Group size: 1 to 3 students.","http://www.sensornets.csiro.au/frontpage.htm","","",""
"SK003","Multi-hop Aerial Wireless Networks","SK0","","","","Active","CECSSE","R & D","","Deploying a large scale wireless sensor network consisting of micro-air vehicles (MAVs) for applications such as bushfire monitoring and control has become possible due to the availability of low cost and ease of manoeuveraiblity of MAVs. The research community has made some progress in understanding the challenging area of multi-hop wireless network of stationary nodes in recent years. However, building aerial multi-hop wireless networks still remains a vision today. In this project you will implement a multi-hop aerial network consisting of MAVs equipped with wireless sensor nodes and GPS devices. There is scope for multiple thesis projects within this umbrella project. Examples include: (i) Understanding Wireless Radio Frequency (RF) behaviour in air-to-air communication as well as air-to-ground communication, (ii) Implementing a distributed coverage algorithm wherein the MAVs will navigate and fill coverage holes created in the network and (iii) designing an automated algorithm for detecting faults in the network. The projects will require students to program wireless sensor nodes (extensive resources and support are available in the Networks Research Lab). The sensor nodes run an operating system called TinyOS and most programs are written in NesC (a variant of C). The Networks Research Lab has over 100 sensor nodes and a few MAVs, which will be used for testing. Thesis students will get to interact with PhD students and researchers in the Networks Lab.","","COMP3331","","Good programming skills"
"SK100","Experimental Characterisation of a Wireless Multi-hop Aerial Wireless Sensor Network","SK","Salil Kanhere","","","Active","CECSSE","R & D","","Wireless sensor networks present significant technological and business opportunities due to their potential to embed sensing and communication in a wide variety of environments. Each wireless sensor node is equipped with computation power, miniature sensors (e.g.: temperature, carbon dioxide) and a wireless radio. Thanks to the advances in semiconductor technology and Micro-Electro-Mechanical system production, these wireless sensor nodes have a size comparable to a match box at present but are likely to diminish further, and can be manufactured cheaply on a mass scale. These devices are particularly suited for remote collection and monitoring of both scientific and business data, due to their ability to form a multi-hop communication network. Consequently, sensor networks are increasingly being used in a variety of applications ranging from environmental monitoring and healthcare to industrial automation.This project seeks to amalgamate the sensing and communicating capabilities of wireless sensor networks with the autonomous flying ability of Micro Aerial Vehicles (MAV) to engineer a novel paradigm, Multi-hop Aerial Wireless Sensor Networks (MAWSN), wherein unmanned aerial vehicles equipped with wireless sensing devices sample the physical space in three dimensions and relay the data over the underlying multi-hop wireless mesh network. The Networks Research group has access to a few MAVs and several sensor nodes, which will be used to build a testbed. This test bed will be used to conduct several experiments to characterise the air-to-air and air-to-ground multi-hop wireless links. ","","COMP 3331 Computer Networks and Applications. Good Programming Skills.","",""
"SK101","Automatic collection of fuel prices from a network of mobile cameras","SK","Salil Kanhere","","","Active","CECSSE","R & D","","It is an undeniable fact that people want information. Unfortunately, even in todayÕs highly automated society, a lot of the information we desire is still manually collected. An example is fuel prices where websites providing fuel price information either send their workers out to manually collect the prices or depend on volunteers manually relaying the information. The development will be carried out on Nokia N95 mobile phones using Symbian OS.This project proposes a novel application of wireless sensor networks to automatically collect fuel prices from camera images of road-side price board (billboard) of service (or gas) stations. The system exploits the ubiquity of mobile phones that have cameras as well as users contributing and sharing data. In our proposed system, cameras of contributing users will be automatically triggered when they get close to a service station. These images will then be processed by computer vision algorithms to extract the fuel prices.","","COMP3331, good programming background.","",""
"SM","Comparison of Queryable Text Compressors","SM","Sebastian Maneth","","","Inactive","BINFBIOMCECSSE","R & D","","A compressor is called ""queryable"" if it is possibleto execute a query on the compressed output, withoutprior decompression. Most of the well known compressiontools such as gzip, bzip2, etc are not queryable.However, there exists a group of comressors which allowquery execution on their output: grammar-based compressors.The output of the latter type of compressor is a context-freegrammar that generates the given input. In this project we want to design a benchmark for testingvarious queryable compressors against each other.The main aspects of the benchmark are (1) to assess thecompression performance of the different compressors interms of their compression ratios and (2) to assess theirprocessing speed when executing queries on their compressedoutput.","","Good programming skills.","",""
"SM03","XML File Compression Tool","SM","Sebastian Maneth","","","Active","BINFBIOMCECSSE","R & D","","Similar to gzip (which compresses well on text files) a compression tool particularly suited for XML files shall be implemented. The idea is to compress data values in the XML file using conventional methods (for text values, e.g., gzip's method) and to compress the tree structure of the XML document using a novel method of ""grammar-based tree compression"". In order for the compressor to run time and memory efficient, the BPLEX tree compression algorithm has to be adapted to work in a streaming fashion. For further details of this project, please contact Sebastian Maneth (smaneth@cse).","","Good programming skills.","",""
"SM04","Novel Tree Compression Algorithm","SM","Sebastian Maneth","","","Active","BINFBIOMCECSSE","R & D","","In 1999, Larsson and Moffat introduced the dictionary-based compression scheme ""RE-PAIR"",  for compressing strings (sequences of letters). The idea in RE-PAIR is that of ""recursive pairing"": Given a string, (1) identify symbols a and b such that the pair ab is the most frequent pair of adjacent symbols in the string (if no pair appears more than once, then stop); (2) introduce a new symbol A and replace all occurrences of ab with A; (3) Repeat from step (1).In this project we want to generalize the RE-PAIR compression scheme from strings to ordered trees.Given a tree, the generalized RE-PAIR algorithm will produce a small tree grammar, which represents the given tree. It will be interesting to compare the compression rates achieved with the new tree RE-PAIR algorithm against the grammar-based tree compression algorithm BPLEX.For more details contact Sebastian Maneth (smaneth@cse).","","Good programmer. Good in alrgorithms and data structures.","",""
"SM06","Benchmark Tool for Testing XML Memory Mappings","SM","Sebastian Maneth","","","Active","BINFBIOMCECSSE","R & D","","An XML Memory Mapping (xmm) maps the content of an XML document into a representation in main memory.The Performance of an xmm is determined by two factors (1) the SIZE of the representation in terms of used memory (2) the TIME needed to access the representation. In this project a benchmark tool shall be programmed, which tests the size and time behavior of different xmms. In order to measure the access time of an xmms, different access methods shall be tested; among those: (1) DOM access(2) top-down access(3) real queries (hand coded, or using a translation algorithm that is particular to the xmm)(4) XQueries (using a proxy of the xmm)For each of the four access methods, appropriate benchmark traversals should be constructed, which can be executed on arbitrary XML input. Furthermore, if time permits, also access times for write access methods should be tested, in particular, simple relabelings/removals, simple replacements/insertions, arbitrary XUpdates.For further details of this project, please contact Sebastian Maneth (smaneth@cse).","","Good programmer. Familiar with XML.","",""
"SM07","XPath Benchmark Tool","SM","Sebastian Maneth","","","Active","BINFBIOMCECSSE","R & D","","XPath is a simple query language for locating nodes in an XML document (tree). XPath is a W3C standard that is widely accepted by industry; it provides the basis of several other, more complex, query languages such as XQuery. While by now there exist several generators for XML benchmark data, like XMark, there are few or none generators of XPath benchmark queries. Recent research papers propose to distinguish 7 or 8 different basic types of XPath queries which inhibit orthogonal complexities of XPath. Based on those and similar observations, a tool shall be programmed which (randomly) generates an arbitrary amount of XPath queries, which can be executed on a specified XML document. The generator should be parameterized by values for the different ""complexity ingredients"".For further details of this project, please contact Sebastian Maneth (smaneth@cse).","","Good programmer. Familiar with XML.","",""
"SM08","Invertible Tree Transformations","SM","Sebastian Maneth","","","Active","BINFBIOMCECSSE","R & D","","XML is often too verbose for human use. Therefore, many people use XSLT stylesheets to convert from XML syntax to an alternative syntax. However, such transformations are not reversible and no general tool exists to automatically parse the alternative syntax back into XML. Based on this observation Brabrand/Moller/Schwartzbach invented XSugar, a tool for managing dual syntax for XML that is based on context-free grammars. In this project we would like to extend the work of Brabrand/Moller/Schwartzbach to a larger classes of translations. The basic underlying questions here is of theoretical interest too: for which large class C of tree transformations is it decidable whether or not a given translations is reversible; ""reversible"" means that the inverse of the translations is a member of the same class C of transformations. In this project several classes of tree transformations known from tree language theory shall be investigated wrt the question of decidable reversibility. In the second part of the project - if time permits - a decision procedure should be implemented for a large class C, that has been identified in the first part.For further details of this project, please contact Sebastian Maneth (smaneth@cse).","","Very good mathematical skills. Familiar with formallanguages and automata theory. Interest in theoretical computer science.","",""
"SM09","Comparison of Queryable Text Compressors","SM","Sebastian Maneth","","","Active","BINFBIOMCECSSE","R & D","","A compressor is called ""queryable"" if it is possibleto execute a query on the compressed output, withoutprior decompression. Most of the well known compressiontools such as gzip, bzip2, etc are not queryable.However, there exists a group of comressors which allowquery execution on their output: grammar-based compressors.The output of the latter type of compressor is a context-freegrammar that generates the given input. In this project we want to design eka benchmark for testingvarious queryable compressors against each other.The main aspects of the benchmark are (1) to assess thecompression performance of the different compressors interms of their compression ratios and (2) to assess theirprocessing speed when executing queries on their compressedoutput.","","Good programming skills.","",""
"SM10","Grep on Compressed Texts","SM","Sebastian Maneth","","","Active","BINFBIOMCECSSE","R & D","","There is a trend in computer science to design compressedformats such that it is possible to execute algorithms directly on the format, without prior decompression.Here we want to design one particular algorithm to runon compressed formats: we want to implement a grep toolthat works on compressed texts. First we will focus oncompressors which generate context-free grammars. Theseare particularly well suited for running grep or otherqueries. Second and more challenging, we would like todevelop a grep that works on gzipped output; this willmost likely require block-wise decompression, but shouldstill outperform the gunzipping followed by the grep.","","Very good programming skills in C/C++.","",""
"SMP","Mutual impact of concurrently executed applications on multiprocessor architectures ","SMP","Stefan Petters","","","Active","CECS","Research","","ne of the bottle necks on multiprocessor platforms is contention for the shared bus (primarily memory, but also others).Within this work we try to quantify the temporal impact of different applications on each other. Namely how the execution time of a given application is affected by the concurrent execution of more or less memory bound applications on another CPU-core. The analysis will be largely based on measurements of time and performance management counters (PMC) of a given workload with other well defined workloads on the other core. The work implies the following:- Modifying the linux SMP code to be able to run well defined sample applications at a given time on a CPU core. This may be achieved using RTAI or a similar linux application.- Once the test bed is established, running various applications concurrently and recording time and various PMC readings.- Then the data needs to be statistically analysed to find the best PMCs for making predictions of impact of applications on each other (experience how to best do this is here in house based on power management work).- Develpoping and instantiating a model of the impact (experience how to best do this is here in house based on power management work). ","","COMP3231","","C/C++ programming"
"SMP101","C++ source code analysis for real-time systems ","SMP","Stefan Petters","","","Active","All Students","Research","","An essential step in the estimation of the worst case execution times of software, is the establishment of a control-flow-graph of the software. This can be either achieved by analysing the source or the object code or the traces provided by a program execution. However, neither analysis alone yields the correct picture. This source code provides only a rough picture, due to the results of optimisations performed by the compiler, while many issues like, for example, indirect branches and calls, can not be resolved by looking at the object-code alone. The analysis of traces is unsafe as only observed transitions between basic blocks will be considered.In order to apply a worst case execution time analysis on L4 micro kernel code, a parser of a subset of C++ code is essential. This should establish the control flow graph of the source code and yield additional information, which is otherwise hard to establish from the object code, like loops with multiple exit or continue conditions, bounds on the number of loop iterations were possible, possible jumps/calls to multiple destinations in the code. This step will also be used to automatically introduce instrumentation into the source code.","http://www.disy.cse.unsw.edu.au/","","",""
"SMP103","Fusion of Several Control Flow Graphs","SMP","Stefan Petters","","","Active","All Students","Research","","Structural analysis of code is one of the major steps for the real-time analysis of code. There are several approaches to this problem, but none of these is ultimately right. The structural analysis of source code, for example, suffers from optimisations done by the compiler, which might change significantly the structure of the code. Object code analysis, in turn, has considerable problems of identifying indirect function calls. Within this project a mechanism to fuse the information available in several control flow graphs available in an XML file has to be designed and implemented. This requires some pattern matching and rules how to integrate conflicting information. This project will be in close collaboration with a senior researcher in the NICTA ERTOS program.","","","",""
"SMP104","Coordination Instance for a Detailed Execution-Time Analysis ","SMP","Stefan Petters","","","Active","All Students","Research","","One aspect of the estimation of the longest execution time of real-time software is the influence of the implementation of a given algorithm. Often small changes in the implementation can result in major changes in the overestimation. This project is focussed on identifying possible sources of overestimation and inform the user, where in his code, hand optimisation looks most promising. Typical examples are code, which is only executed in the first iteration of a loop and thus lead to severe overestimations, or small if-then-else constructs, which may be recoded to allow for larger units to be analysed. Supervision will be directly by a Senior Researcher.","","","",""
"SMP105","A Real-Time Programmers Guide Dog towards Better PredictableSoftware","SMP","Stefan Petters","","","Active","All Students","Research","","One aspect of the estimation of the longest execution time of real-timesoftware is the influence of the implementation of a given algorithm.Often small changes in the implementation can result in major changes inthe overestimation. This project is focussed on identifying possiblesources of overestimation and inform the user, where in his code, handoptimisation looks most promising. Typical examples are code, which isonly executed in the first iteration of a loop and thus lead to severeoverestimations, or small if-then-else constructs, which may be recodedto allow for larger units to be analysed. Supervision will be directlyby a Senior Researcher.","","","",""
"VT01","Analysis of Methods Improving IT Support for Business","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","Research","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Several research communities (e.g., value-based software engineering, business-driven IT management, IT governance) have emerged to address various aspects of business-IT alignment. Since all these communities work on the problem of business-IT alignment, there are some similarities and overlaps between them. However, there are also important differences, caused by the diversity of research community backgrounds, methodologies, and focus topics. In addition, different works from the same research community often cover different business aspects or the same aspects in different ways.    Students working on this thesis project will read a number of recent papers presenting different software and/or system engineering methods to make IT better aligned with business. They will critically analyze and compare results from different authors (possibly, across different research communities). The expected result of their work is better understanding of which business aspects (e.g., business goal, business value, business strategy, business process, business context, etc.) are supported by different methods and to which extent. Furthermore, the students will synthesize recommendations which existing methods to use in engineering and management of IT systems that have to be aligned with businesses in different ways, as well as recommendations of how to extend the existing methods to improve their coverage of different business aspects.    There will be no development (e.g., implementation), in this research-only topic. However, the knowledge that the students will gain from this thesis project will be very useful in their IT industry jobs after graduation (including software engineering and system administration jobs). ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html).    The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT02","Integrating Value-Based Software Engineering (VBSE) and Business-Driven IT Management (BDIM)","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","R & D","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Value-based software engineering (VBSE) and business-driven IT management (BDIM) both try to make IT systems better aligned with business, but research different problems and take different approaches. VBSE explicitly considers value issues (e.g., value-based prioritization) during software engineering, in order to make the resulting systems more useful. For example, it tries to determine software requirements that best support organizationÕs business strategy. On the other hand, the goal of BDIM is to determine mappings between technical and business performance metrics and leverage them to make run-time IT system and/or service management (monitoring, control) decisions that maximize business value. For example, it tries to quantify impact on business profits of increased/decreased availability or response time of IT systems and services.    The VBSE-BDIM integration should research how to bi-directionally transfer business value information between design-time and run-time artifacts (e.g., models) developed by both communities and how to use this information in novel closed-control loops. This includes finding how to best: a) engineer IT systems during design-time to maximally support run-time BDIM activities, b) manage IT systems/services during run-time to align with and leverage business value information captured in design-time VBSE models, and c) feed back run-time BDIM information about business value compliance to improve VBSE design-time models.    Students working on this thesis project will first survey the currently leading approaches to VBSE, BDIM, and their integration. Then, they will design, implement, and test extensions of popular open-source tools for software engineering (e.g., Eclipse) or distributed system provisioning (e.g., Apache Axis2, ActiveBPEL) to support one approach to VBSE-BDIM integration. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT03","Value-Based Software Engineering of Web Services and Business Processes","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","R & D","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Value-based software engineering (VBSE) is one approach to make IT systems better aligned with business. It explicitly considers value issues (e.g., value-based prioritization) during software engineering, in order to make the resulting systems more useful. For example, it tries to determine software requirements that best support organizationÕs business strategy.    Web services are distributed computing application components that use a number of Extensible Markup Language (XML) based technologies to implement the service-oriented architecture (SOA). Implementation-independence of Web services technologies allows different businesses to collaborate and achieve common business goals despite the fact that the collaborating Web services can be distributed over the Internet, run on different platforms, and implemented in different programming languages. Web services technologies are already embedded in various products and services of all major computing companies and used for diverse purposes. One of the most important application areas of Web services is e-business process integration in business-to-business (B2B) and/or enterprise application integration (EAI) scenarios. Web services are also increasingly used for business process management.    Students working on this thesis project will first survey the currently leading approaches to VBSE, with particular emphasis on how these approaches suit (or do not suit) development of business systems implemented with Web services. Then, they will design, implement, and test extensions of popular open-source tools for software engineering (e.g., Eclipse) or Web service orchestration into business processes (e.g., ActiveBPEL) to support one approach to value-based software engineering of Web services and business processes. Based on this experience, they will also synthesize recommendations of how to improve the current VBSE approaches. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT04","Business-Driven IT Management of Web Services and Business Processes","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","R & D","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Business-driven IT management (BDIM) is one approach to make IT systems better aligned with business. Its goal is to determine mappings between technical and business performance metrics and leverage them to make run-time IT system and/or service management (monitoring, control) decisions that maximize business value. For example, it tries to quantify impact on business profits of increased/decreased availability or responsiveness of IT systems and services.    Web services are distributed computing application components that use a number of Extensible Markup Language (XML) based technologies to implement the service-oriented architecture (SOA). Implementation-independence of Web services technologies allows different businesses to collaborate and achieve common business goals despite the fact that the collaborating Web services can be distributed over the Internet, run on different platforms, and implemented in different programming languages. Web services technologies are already embedded in various products and services of all major computing companies and used for diverse purposes. One of the most important application areas of Web services is e-business process integration in business-to-business (B2B) and/or enterprise application integration (EAI) scenarios. Web services are also increasingly used for business process management.    Students working on this thesis project will first survey the currently leading approaches to BDIM, with particular emphasis on how these approaches suit (or do not suit) business systems implemented with Web services. Then, they will design, implement, and test extensions of popular open source tools for Web service provisioning (e.g., Apache Axis2) or Web service orchestration into business processes (e.g., ActiveBPEL) to support one approach to business-driven IT management of Web services and business processes. Based on this experience, they will also synthesize recommendations of how to improve the current BDIM approaches. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT05","Tools for Value-Based Software Engineering","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","Development","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Value-based software engineering (VBSE) is one approach to make IT systems better aligned with business. It explicitly considers value issues (e.g., value-based prioritization) during software engineering, in order to make the resulting systems more useful. For example, it tries to determine software requirements that best support organizationÕs business strategy.    Students working on this thesis project will design, implement, and test extensions of popular open-source tools for software engineering (e.g., Eclipse) to support one approach to value-based software engineering. This is a development thesis project, where the focus is on implementation and hands-on experiments, instead of surveys and analysis of research literature. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT06","Tools for Business-Driven IT Management","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","Development","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Value-based software engineering (VBSE) is one approach to make IT systems better aligned with business. It explicitly considers value issues (e.g., value-based prioritization) during software engineering, in order to make the resulting systems more useful. For example, it tries to determine software requirements that best support organizationÕs business strategy.    Students working on this thesis project will design, implement, and test extensions of popular open-source tools for Web service provisioning (e.g., Apache Axis2) or Web service orchestration into business processes (e.g., ActiveBPEL) to support one approach to business-driven IT management. This is a development thesis project, where the focus is on implementation and hands-on experiments, instead of surveys and analysis of research literature. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT07","Business-Driven Selection of Web Services","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","R & D","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Web services are distributed computing application components that use a number of Extensible Markup Language (XML) based technologies to implement the service-oriented architecture (SOA). Implementation-independence of Web services technologies allows different businesses to collaborate and achieve common business goals despite the fact that the collaborating Web services can be distributed over the Internet, run on different platforms, and implemented in different programming languages. Web services technologies are already embedded in various products and services of all major computing companies and used for diverse purposes.    The Universal Description, Discovery and Integration (UDDI, http://www.uddi.org/) is an industrial standard for describing businesses and their Web services and allows discovery and selection of Web services during run-time. This supports flexible and adaptive run-time composition of Web services to address changing business needs. However, UDDI describes only ÒwhatÓ Web services do and not Òhow wellÓ they do it. As the number of Web services that offer similar functionality increases in the global market, comprehensive description of their quality of service (QoS, such as response time, availability) and various metrics of business value (e.g., price, price/performance ratio, customer satisfaction) will become major competitive advantages. In many cases, Web services will provide the same or very similar operations, but with different QoS and business value. While several UDDI extensions specifying QoS information appeared recently, specification of business value is still relatively superficial.    Students working on this thesis project will extend an existing open-source UDDI implementation with capabilities for publication and search of various business value measures (and, possibly, other relevant information, such as QoS). They will also develop novel algorithms for selection among similar Web services to maximize various business value metrics. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT08","Extending Service-Oriented Middleware for Business-Driven IT Management","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","Development","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Web services are distributed computing application components that use a number of Extensible Markup Language (XML) based technologies to implement the service-oriented architecture (SOA). Implementation-independence of Web services technologies allows different businesses to collaborate and achieve common business goals despite the fact that the collaborating Web services can be distributed over the Internet, run on different platforms, and implemented in different programming languages. Web services technologies are already embedded in various products and services of all major computing companies and used for diverse purposes.    Management of IT systems (including service-oriented systems) is the process of their monitoring and control to ensure correct operation, discover and fix problems (e.g., faults, performance problems), and achieve maximal benefits from the IT systems. Ideally, IT systems should be self-managing, or at least be managed with minimal human intervention. This is one of the goals of autonomic computing. The current management solutions for service-oriented systems focus on optimizing technical quality of service (QoS, e.g., response time and availability), but provide inadequate treatment for optimizing business metrics, such as profit and customer retention.    Students working on this thesis project will design, implement, and test extensions of an existing autonomic service-oriented middleware (already measuring technical quality of service) with additional software support (implementing new algorithms, data structures, XML formats) for maximization of business metrics. This is a development thesis project, where the focus is on implementation and hands-on experiments, instead of surveys and analysis of research literature. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT09","Extending UML Modeling Tools for Specification of Business Value Information","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","Development","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.     The Unified Modeling Language (UML) became a widely used industry standard for specification of design-time software models. Unfortunately, UML does not enable specification of very important design information related to expected business value (e.g., price, price/performance ratio, customer satisfaction) and desired quality of service (QoS, such as response time, availability).    Students working on this thesis project will design, implement, and test new extensions of an existing popular open-source UML modeling tool to enable specification of business value metrics (and, possibly, other relevant information, such as QoS) in UML models. An emphasis will be put on UML models of service-oriented architectures (SOAs). For example, the test examples will be UML models of various Web service systems. If time allows, the students might also design, implement, and test novel algorithms for exporting/importing the resulting annotated UML models and for processing of these exported files. This is a development thesis project, where the focus is on implementation and hands-on experiments, instead of surveys and analysis of research literature. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT10","Extending BPMN Modeling Tools for Specification of Business Value Information","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","Development","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.     The Business Process Modeling Notation (BPMN) is an industry standard for specification of business process models. Its popularity rapidly increases. Unfortunately, BPMN does not enable specification of very important information related to expected business value (e.g., price, price/performance ratio, customer satisfaction) and desired quality of service (QoS, such as response time, availability).    Students working on this thesis project will design, implement, and test new extensions of an existing popular open-source UML modeling tool to enable specification of business value measures (and, possibly, other relevant information, such as QoS) in BPMN models. An emphasis will be put on BPMN models of business processes implemented with service-oriented architectures (SOAs), such as Web service systems. If time allows, the students might also design, implement, and test novel algorithms for exporting/importing the resulting annotated BPMN models and for processing of these exported files. This is a development thesis project, where the focus is on implementation and hands-on experiments, instead of surveys and analysis of research literature. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT11","Translation between Different Business Value Models","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","R & D","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Several research projects have emerged to address various aspects of business-IT alignment. While the concept of business value to be maximized is very important (or even central) to these projects, there is no widely accepted definition of its precise meaning and every projects provides their own business value models. In this thesis project, by Ôbusiness valueÕ we will mean any measure of worth of a business entity. This is a broad concept that includes financial aspects (e.g., earned income, costs, profit), but also a number of other aspects (e.g., number of customers, market share, customer satisfaction) that are important for business operations.    Students working on this thesis project will study several recent projects that model business value of IT systems and develop a new general model that is a superset of the existing models. This general model should have a representation in the Extensible Markup Language (XML). In addition, the students will also design, implement, and test algorithms for translation (using XML processing) between the existing business value models and their new general model. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"VT12","Ontology of Business Value","VT","Vladimir Tosic","","","Active","BINFBIOMCECSSE","Research","","   To maximize benefits from information technology (IT) systems, their execution should be coordinated with business goals, business value measures, business strategies, and business processes of user organizations. Unfortunately, improvements in IT (e.g., lower response time, higher availability) need not result in expected business benefits (e.g., profit, customer satisfaction), often because of poor alignment between business operations and used IT systems.    Several research projects have emerged to address various aspects of business-IT alignment. While the concept of business value to be maximized is very important (or even central) to these projects, there is no widely accepted definition of its precise meaning and every projects provides their own business value models. In this thesis project, by Ôbusiness valueÕ we will mean any measure of worth of a business entity. This is a broad concept that includes financial aspects (e.g., earned income, costs, profit), but also a number of other aspects (e.g., number of customers, market share, customer satisfaction) that are important for business operations.    In computer science (as opposed to philosophy), an ontology is a formal data model of a group of concepts within some domain, along with relationships between these concepts. For example, an ontology of a university would model concepts such as students, lecturers, courses, classrooms, research grants, etc., as well as relationships such as that lecturers give courses and students attend courses. An ontology differs from other data models (e.g., those that can be developed with entity-relationship diagrams) in the use of logic-based formalisms that enable powerful queries, analyses, and logical reasoning about objects with the domain represented with this ontology. Ontologies are essential to a number of new computing (particularly artificial intelligence) technologies, such as Semantic Web that aims to make that software can more easily process content published on the Web (the current Web was made for navigation by humans and due to a number of ambiguities its content cannot be easily processed by software).    Students working on this thesis project will study several recent projects that model business value of IT systems and will develop an ontology of concepts (and their relationships) within and across these works. In effect, they will develop a new general model of business value that is a formalized superset of the existing models. To represent this ontology, the students will use a popular ontology language (used in the Semantic Web) and a popular open-source tool.    This thesis topic is classified as research-only, because the emphasis is on literature search and analysis, as well as (theoretically grounded) synthesis of an ontology. While the thesis project will require representation of this ontology in a particular ontology language and with a particular tool, this development aspect is secondary (it can be viewed as part of validation of the design/synthesis of this ontology). However, the knowledge that the students will gain from this thesis project will be very useful in their IT industry jobs after graduation, because of the omnipresence of the problem of business-IT alignment and the importance that Semantic Web will have in the future. ","","","","   The students will work closely with researchers at NICTA (http://www.nicta.com.au) in a friendly team environment led by Prof. Ross Jeffery (http://www.cse.unsw.edu.au/db/staff/info/rossj.html). For further information, email: Dr. Vladimir Tosic (ÔvtosicÕ at the CSE e-mail system) with Subject line ÒUNSW ThesisÓ. "
"WEW08a","Enterprise Search Engine","WEW","Wei Wang","","","Active","BINFBIOMCECSSE","R & D","","While Web search engines have been a huge success, very poorenterprise search is the norm. Enterprise search presents several unique challenges that prohibitimmediate application of search techniques developed in the domain ofWeb search:(1) Search heteogeneous repositories. The system needs to crawl overHTML/XML pages, databases, CMSs, spreadsheet and word processingdocuments, emails, and legacy systems.(2) Search result quality. On one hand, it is not easy to rankheteogeneous items in the search result. On the other hand, there aremuch fewer (good-quality) anchors/links among objects and this makesPageRank-based ranking algorithm inapplicable. (3) The need to relate information together. A unique requirement isto assemble related information together and present them in anintuitive way to the user. E.g., a search for a customer's name shouldbring out her contact details from some electronic forms and her pastpurchase records from a database. (4) Enterprise-specific requirement. E.g., search result rankingprobably should be tuned to represent enterprise's specific goals (sothat visitors get to see what the enterprise want them to see). Queryrefinement/expansion should reflect the local knowledge of theenterprise (e.g., GIVE is the assignment submission system used inCSE, not just a verb). Another common requirement is to find peoplewith a particular expertise in a large enterprise. Security and fineaccess control are also very important demands. In this project, we are going to build an extensible enterprise searchengine based on some open-source toolkits (e.g., nutch and lucene). Wewill take CSE Web pages and databases as the first step. References:[1] David Hawking: Challenges in Enterprise Search. ADC 2004: 15-24.[2] http://www.microsoft.com/enterprisesearch","","COMP3311","","- Enthusiastic about Search Engines- Good programming skills- Knowledge in Data Mining, Machine Learning, or Artificial Intelligence is a plus"
"YS20","Developing an FPGA-based Smart Camera for Car Tracking and Counting","YS","David Shi","","","Active","CECS","Development","","Aim of research is to develop a smart camera prototype using FPGA platform. The camera will detect, analyse, track vehicle movement and count them.Different from desktop and PC-based image and video processing, building a camera from scratch using FPGA as central processor is a challenging piece of work. It allows student(s) to learn state-of-the-art embedded system technologies. Student(s) will learn to use hardware description language to program FPGA. Implement algorithms already developed by researchers. Extensive coding, debugging and testing.Smart camera can perform detection and decision-making all by itself, without having to send video to remote PC for processing. Developing image processing algorithms suited for embedded environment without sacrificing performance.The outcome of the project will be a smart camera as an FPGA-based embedded system that can detect and track vehicle movement on the roads and count them. ","","Basic image processing knowledgeInterest and knowledge in FPGA and hardware programming","",""
"YS21","Developing and implementing algorithms for video-based hand tracking for human computer interface applications. ","YS","David Shi","","","Active","CECS","R & D","","This project aims to develop and implement, under guidance by researchers, hand-tracking algorithms using either skin-color segmentation, or shape and motion based segmentation. Applications include human computer interfaces. Students will research into different types of algorithms, test and select algorithms based on application requirements. Coding implementation, test and debug. Language can be C, C++, Java, or VHDL.Main contribution will be to develop hand-tracking function for a camera that can recognise computer userÕ hand gestures. Working algorithms satisfying design requirements. Building prototype and demonstrator.","","Signal processing and image processing experiences","",""
"YS22","Developing an FPGA-based Smart Camera development platform. ","YS","David Shi","","","Active","CECS","Development","","An FPGA is a high performance computing platform widely used in networking, image/video processing and other high processing power demanding applications. This project aims at researching into efficient ways of integrating various image processing modules into an FPGA based intelligent camera platform. The project will involve HDL and C/C++ coding, debugging and real-time testing.Working alongside a senior researcher, the student will have a great opportunity to gain skills in advanced FPGA technologies and computer/machine vision, he/she will also learn how to do research on real-time video processing and build an intelligent video camera and gain insight into embedded imaging system. A smart camera is a vision system in which the primary function is to produce a high-level understanding of the imaged scene and generate application-specific data to be used in an autonomous and intelligent system. The reason a smart camera is ÔsmartÕ lies in the fact that there exists inside the camera a processing unit which performs application specific information processing. Using an FOGA to build a smart camera, from scratch, is a challenging piece of work. The primary function of the camera to be built Ð gesture recognition Ð has a wide range of applications. The outcome of the project will be both a smart camera development platform and a camera prototype that can track simple moving objects. ","","Image processing experienceInterest and knowledge in FPGA and hardware","",""
"YS23","Stress and cognitive load analysis for HCI (Human Computer Interaction) Applications","YS","David Shi","","","Active","CECS","R & D","","Physiological measurements such as skin conductance, heart rate and blood pressure can be used as part of usability evaluation framework for HCI (human computer interfaces). These measurements can provide indication as to whether a user is under stress when completing a task using a computer or application.  The GSR (Galvanic Skin Response) is one type of physiological measurement that can be used to indicate whether a user is under stress compared to relaxed state, and can potentially indicate userÕs cognitive load levels.  This project aims to validate the above hypotheses/statement. Main tasks include understanding of GSR, designing user studies, developing real-time automated GSR analysis tools using Java or a script language.The design of a user study to validate and test several hypotheses is a research process, which will be carried out with help from researchers. The user study will be carried out by the student with help from researchers. And a tool set to automate the GSR data acquisition, real-time display and analysis will be completed by the student.Details:-         Learn some basics about physiological sensors, and especially about GSR-         Understand the goals and requirements of user study and analysis tools-         Design user study with the goal of designing tasks that can produce different levels of cognitive loads and especially can induce different types of stresses.-         Develop s/w modules for GSR acquisition, visualisation, analysis-         Run user study-         Data analysis-         Document the design and implementation-         Run a demonstration, prepare report and give a seminar presentation","","","",""