Research


Here you will find information on what research problems I am working on, what results we have got so far and links to some of my publications.

Multi-rate wireless mesh networks

This research is supported by an Australian Research Council Discovery Grant. This work is done in conjunction with Archan Misra (IBM Research) and Junaid Qadir (PhD student, UNSW).

Project synopsis: In a multi-rate wireless network, a node can dynamically adjust its link transmission rate by switching between different modulation schemes. For the current IEEE802.11a/b/g standards, this rate adjustment is limited to unicast traffic. In this project, we study the problem of efficient routing and packet distribution for multicast traffic flows in a multi-rate wireless mesh networks (WMN). We assume that the MAC layer of future WMNs will provide some form of multicast support, where the transmitter may be able to specify the transmission rate of the MAC-layer multicast, and, either explicitly or implicitly, the recipients of the multicast. Assuming such MAC layer multi-rate multicast capability, our goal is to study how low-latency (and possibly high throughput) network layer multicast of data traffic can be realized.

Research results: We consider two different cases
  1. Single-radio single-channel multi-rate WMN:
  2. Multi-radio multi-channel multi-rate WMN:
Broadcast in single-radio single-channel multi-rate WMN:

Our key research questions are:

  1. Effect of Multi-Rate Links on Efficient Broadcasting: Is multi-rate multicast at the link layer necessary for realizing a low broadcast latency? By what factor can the introduction of multi-rate multicast reduce the broadcast latency compared with single-rate multicast?
  2. Choice of transmission rates in multi-rate networks: If multi-rate multicast is to be introduced, how many different transmission rates do we need? How should they be chosen? Are some rates more efficient than others?
In order to answer these questions, we formulate the minimum broadcast latency optimisation problem and derived heuristics to solve it. The heuristics exploits the multi-rate feature of multi-rate WMNs as well as wireless multicast advantage. By using these heuristics and simulations based on current IEEE 802.11 parameters, we show that multi-rate multicast can reduce broadcast latency by 3-5 times compared with using the lowest rate alone.

Our first heuristic (BIB, based on modifying Prim's algorithm) was published in "Low latency multimedia broadcast for multi-rate wireless meshes" in WiMesh'05 (Invited Paper).

Subequently, we have come out with an improved heuristics (WCDS based on weighted connected dominating set).

These algorithms allow us to study the second set of research questions posed above. Our key finding is that the efficiency of a transmission rate for broadcast can reasonably be predicted by the product of the transmission rate and its transmission coverage area (or rate-area product or RAP for short).

The WCDS algorithm and the concept of RAP are described in: "Low latency broadcast in multi-rate wireless mesh networks", IEEE JSAC Special Issue
in Multi-hop Wireless Mesh Networks (to be published in the 4th quarter of 2006).

All the above-mentioned results are described in details in the technical report "Low latency broadcast in mult-rate wireless mesh networks" (UNSW Computer Science and Engineering Technical Report No. UNSW-CSE-TR-0514).

Here is a presentation which summarises the above results: "Multi-rate multicast in wireless mesh networks".

Broadcast in multi-radio multi-channel multi-rate WMN:

The key research question is to study how to perform low latency (network layer) broadcast in a multi-radio multi-channel multi-rate WMN. We found that the broadcast latency is minimised by exploiting rate-diversity, interface-diversity and wireless multicast advantage.

The results can be found in the technical report: "Minimum Latency Broadcasting in Multi-Radio Multi- Channel Multi-Rate Wireless Mesh Networks" (
UNSW Computer Science and Engineering Technical Report no. UNSW-CSE-TR-0608)

A shorter version of this report will appear in SECON 2006.

Top of the page


Quality of service for wireless mesh networks

The aim of this project is to investigate how quality of service can be provided in wireless mesh networks. This project is funded by the Smart Internet Technology CRC. Here is the project web page.

Other project members are: Sanjay Jha, Salil Kanhere, Joo Ghee Lim, Anjum Naveed and Yang Gao,

Top of the page
Hybrid sensor networks

We consider hybrid sensor networks consisting of both resource-rich and resource-impoverished sensor devices. The resource-rich devices, called micro-servers, are more
expensive but have significantly greater bandwidth and energy capabilities compared to the low-cost, low-powered sensors. Such hybrid sensor networks have the potential
to support the higher bandwidth communications of broadband sensor networking applications, as well as the finegrained sensing that is made possible by smaller sensor devices. Our investigation is focused on the impact of the placement of the resource-rich nodes on the lifetime of the hybrid sensor networks. In particular, we study the lifetime-to-cost ratio of hybrid sensor networks and find that hybrid sensor network is scalable in terms of the benefits that it provides (in terms of increasing lifetime-to-cost) ratio.

The results can be found in "Deploying Long-Lived and Cost-effective Hybrid Sensor Networks". (Accepted by Ad Hoc Networks (Elsevier). Preliminary results appeared in Basenets'04)

Collaborators: Wen Hu, Sanjay Jha (UNSW), Nirupama Bulusu (Portland State University)

Top of the page
Cane toad monitoring

The Cane toad (Bufo marinus) was introduced to Australia in the 1930s in the belief it would control pests in Sugar Cane crops. Since their introduction they have progressively
spread through north-eastern Australia. Their expanding distribution, density and ecology characteristics have raised grave concerns regarding their impact on Australia’s native
fauna.

Our goal is to deploy a large scale, inexpensive wireless sensor network that can operate unattended and is capable of monitoring, tracking and measuring the impact of cane toads in areas such as Kakadu National Park from acoustical observations. It is challenging to implement such a real world sensor network application because of the high sampling rate (10kHz) required and intensive processing required for acoustics samples.

We set up two prototypes of wireless sensor networks that recognize vocalizations of up to 9 frog species found in northern Australia. Our first prototype is simple and consists of only resource-rich Stargate devices. Our second prototype is more complex and consists of a hybrid mixture of Stargates and inexpensive, resource-poor Mica2 devices operating in concert. In the hybrid system, the MICA2s are used to collect acoustic samples, and expand the sensor network coverage.

This work appeared as "The Design and Evaluation of a Hybrid Sensor Network For Cane-toad Monitoring" in IPSN 2005.

Our work on cane toad monitoring was reported in ABC Science online news article: "Tracking cane toads from the sofa"

Click here to go to the project web site.

Collaborators: Wen Hu, Sanjay Jha, Andrew Taylor (UNSW), Nirupama Bulusu (Portland State University)

Top of the page
Video caching

Most proxy caches for streaming videos do not cache the entire video but only a portion of it. This is partly due to the large size of video objects. Another reason is that the popularity of different part of a video can be different, e.g. the prefix is generally more popular. Therefore, the development of efficient cache mechanisms requires an
understanding of the internal popularity characteristics of streaming videos.

We analyze two 6-month long traces of RTSP video requests recorded at different streaming video servers of an entertainment video-on-demand provider, and show that the traces provide evidence that the internal popularity of the majority of the most popular videos obeys a k-transformed Zipf-like distribution. Based on this observation, we propose a caching algorithm which exploits this empirical internal popularity distribution. We find that this algorithm has similar performance compare with fine-grained caching but requires significantly less state information.

This research is patially funded by the Australian Research Council.

Our results can be found in this technical report: "A Dynamic Caching Algorithm Based on Internal Popularity Distribution of Streaming Media"(Technical Report  UNSW-CSE-TR-0515, School of Computer Science and Engineering, UNSW)

The full paper will appear in ACM Multimedia Systems Journal. Earlier results appear in IEEE AINA 2006.

Research student: Yu Jiang (Huazhong University of Science and Technology, China)

Top of the page
Quality of service for MPLS networks

The aim of this project is to consider dynamic routing of restorable Quality of Service (QoS) connections in a multi-protocol label switched (MPLS) network. A restorable QoS connection must be able to provide QoS in both the service and backup paths. The current work focuses on either meeting QoS requirement in the service path alone (known generally as QoS routing) or meeting the resilience requirement alone (i.e. providing service and backup path but without considering QoS). Our work is distinct in the sense that we consider both QoS and resilience simultaneously.
 
Our results are reported in: "Dynamic Routing of Restorable QoS Connections in MPLS Networks" (To appear in LCN'05)

This project is funded by the Smart Internet Technology CRC.

Research student: Filip Rosenbaum (UNSW)

Top of the page
Quality of service for Wireless LAN

We consider the problem of providing relative service differentiation in IEEE 802.11 Wireless LAN by using different Medium Access Control (MAC) parameters for
different service classes. We successfully dervied an analytical model which predicts the saturation throughput of IEEE 802.11 Distributed Coordination Function with multiple classes of service. This model allows us to show that relative service differentiation can be achieved by varying the initial contention window alone. In this case, the saturation throughput of a station can be shown to be approximately inversely proportional to the initial contention window size being used by that station. This provides an easy way for designers to adjust the achievable bandwidth of competing stations.

Results are reported in: "Analytical Model for Service Differentiation Schemes in IEEE 802.11 Wireless LAN" (IEICE Transactions on Communications. Vol E87-B, No. 6, pp. 1724-1729, 2004. Preliminary results appeared in WLN2003.)

Collaborator: Dr. Jianhua He (Bristol University)

Top of the page
Content distribution networking

The current paradigm of Content Distribution Networks (CDN) deployment requires heavy infrastructure investment since a large number of servers have to be deployed over a wide area. To overcome this difficulty, we propose a new paradigm where future CDNs are to be deployed over a leased server infrastructure.

We consider how CDN can be realised in this new paradigm.

Some publications are:
This project is funded by the Smart Internet Technology CRC

Collaborators: Vinh Nguyen, Paul Boustead, Farzad Safaei (University of Wollongong)

Top of the page
ctchou
Last modified: Oct 8 18:05:47 EST 2005