Past ISSNIP Seminars Series

ISSNIP Events Calendar: ISSNIP Weekly Seminars and Group Meetings - Past Seminars 2007

Past Seminars: 2007
Date Time Venue Speaker Title
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne
Ass/Prof. Bhaskar Krishnamachari (Viterbi School of Engineering, The University of Southern California)
Modeling Communication in Wireless Sensor Networks
Abstract: The art of modeling lies in finding simple yet faithful abstractions of reality. I will present some empirically-validated communication models that are tractable enough to be useful in designing, simulating, and mathematically analyzing wireless sensor network protocols operating at the link, network and transport layers.


      Bio: Dr Bhaskar Krishnamachari is Philip and Cayley Early Career Chair Assistant Professor at the Viterbi School of Engineering at the University of Southern California, where he heads the Autonomous Networks Research Group in the departments of Electrical Engineering and Computer Science. He received his B.E. from The Cooper Union for the Advancement of Science and Art in 1998, and his M.S. and Ph.D. from Cornell University in 1999 and 2002 respectively, all in Electrical Engineering. He has co-authored more than 100 technical publications, mostly in the area of wireless networks, including papers that received best paper awards at IPSN 2004 and MSWiM 2006. He received the the U.S. National Science Foundation's CAREER award in 2004, and USC Viterbi School of Engineering's outstanding junior faculty research award in 2005. He served as the vice chair of the sensor networks and ubiquitous computing track at ICDCS 2007, and is the vice chair of the applications track for DCOSS 2008. He is an editor for the Elsevier Journal of Ad Hoc Networks, the Elsevier Journal of Pervasive and Mobile Computing, the ACM Mobile Computing and Communications Review, and the EURASIP Journal of Wireless Communications and Networking. He has authored a book titled Networking Wireless Sensors, published by Cambridge University Press.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne
Prof. Jayadeva (IIT Delhi)

Pervasive Intelligence
Abstract:Talk will include topics such as Twin Support Vector Machines and implementation of SVMs on a chip.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Prof. Hans Weghorn (The University of Stuttgart - Germany)
Efficient Information Access from Constraint Wireless Terminals - Exploiting Personalization and  Location - Based Services
Abstract: Today, the success of data services used from small mobile devices, like digital phones or PDAs, appears very limited. Different reasons can be identified, which prevent the average customer from broadly using wireless data services: At first, the user has to deal with very uncomfortable devices in terms of UI ergonomy, and on the other hand, the costs for wireless data communication are extremely high. These restrictions can be overcome by employing a system concept, which is built up on two main components: A personalized display software allows simplifying the information access on the wireless terminal, while an intermediate agent residing on the Internet takes care of mining the desired contents from the open Web. In addition to the improved UI handling, this concept offers a reduction of costs and an increase in access speed. Real-world experiments with an information system on actual train departures are reported for measuring and demonstrating the benefit of the described system concept.
Green Theatre, Dept of EEE, Building 193, The University of Melbourne
Jayavardhana Gubbi (The University Melbourne)

Algorithms for Protein Structure Prediction
Abstract: Knowledge of Protein Structure is very important in understanding the function of a given protein and in better design of drugs. A novel scheme using molecular replacement for solving the phase problem in protein crystallography will be discussed. The problem is approached by first predicting different attributes of the protein structure - Secondary Structure, Disulphide Bridge, Relative Solvent Accessibility and the fold (Topology). The model is then built using molecular replacement via protein fragments for solving the important phase problem.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Dr Marcelo Espinoza

      Katholieke Universiteit Leuven, Belguim

Structured Kernel Based Modeling for Nonlinear System Identification
Abstract: In the context of nonlinear system identification and forecasting of time-series, important challenges are related to the incorporation of prior knowledge and to the estimation of such models from large scale datasets. In this talk, we start with the Least-Squares Support Vector Machines (LS-SVM) formulations for nonlinear regression, describing the primal-dual optimization framework. This framework can be extended to incorporate structured elements available from prior knowledge about the problem. It will be shown that the prior knowledge becomes part of the kernel function of the model, such that it can be used directly to evaluate the models at new datapoints. This property makes a contribution in terms of modularity of the model formulation, in the sense that different types of prior knowledge can be tested in practice simply by changing the kernel function being used. Furthermore, large scale versions of the different LS-SVM extensions can be formulated in primal space by using the Nystrom method. By considering each of the developed extensions as building blocks, a modular framework for the case of nonlinear system identification is further proposed. Practical examples show the benefits of this approach.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Sutharshan Rajasegarar (Slaven Marusic)

Conference Brief and SensorMap
Abstract:SenseWeb provides a portal for querying and visualising sensor network data in real-time using geographical interface like windows live local. This is to utilise for the dissemination of data from the Great Barrier Reef sensor network deployment. This talk will provide a basic overview of Microsoft tutorial on SensorMap, which was conducted in conjunction with IPSN 2007 conference. We will also give a brief review and pointers to selected papers from that conference.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Patrick Laube (Department of Geomatics, The University of Melbourne)


Happy Sheep - Decentralised Movement Analysis through Local Relational Sensing
Abstract: Advances in location sensor technology offer new reliable and cost-effective means to track moving objects. Being spatio-temporal in nature, movement data tends to be voluminous and hence requires sophisticated analysis techniques to derive high level event knowledge from low level trajectory data. Conventional movement analysis bases on centralised information processing, where global systems such as spatial databases or GIS collate all information in the system and run omniscient data mining algorithms in order to detect salient events. However, when modelling and analysing movement of location-aware roaming agents, such global approaches fail due to the known energy and communication constraints in sensor systems. Hence, new decentralised models are required, where individual moving computing elements infer from locally sensed relational information only whether they are part of a movement pattern. Agents shall not only be 'location-aware' but 'spatially-aware', inferring themselves whether they are on collision course or run into traffic jam. Reviewing recent research on movement analysis, this talk makes a strong argument for the development of techniques for decentralised, in-network detection of salient movement patterns in wireless ad-hoc networks of location-aware moving agents and presents work in progress implementing such decentralised approaches.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Dr Yee Wei Law (Dept. of Electrical and Electronic Eng., The University of Melbourne)

Secure k-Connectivity Properties of Wireless Sensor Networks
Abstract:A k-connected wireless sensor network (WSN) allows messages to be routed via one (or more) of at least k node- disjoint paths, so that even if some nodes along one of the paths fail, or are compromised, the other paths can still be used. This is a much desired feature in fault tolerance and security. k-connectivity in this context is largely a well-studied subject. When we apply the random key pre- distribution scheme to secure a WSN however, and only consider the paths consisting entirely of secure (encrypted and/or authenticated) links, we are concerned with the se- cure k-connectivity of the WSN. This notion of secure k- connectivity is relatively new and no results are yet avail- able. The random key pre-distribution scheme has two im- portant parameters: the key ring size and the key pool size. While it has been determined before the relation be- tween these parameters and 1-connectivity, our work in k- connectivity is new. Using a recently introduced random graph model called kryptograph, we have derived mathematical formulae to estimate the asymptotic probability of a WSN being securely k-connected, and the expected secure k- connectivity, as a function of the key ring size and the key pool size. In this talk, I will give a high-level overview of these concepts and results.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Sutharshan Rajasegarar (Dept. of Electrical and Electronic Eng, The University of Melbourne)

Analysis of Anomalies in IBRL Data from a Wireless Sensor Network Deployment
Abstract: Detecting interesting events and anomalous behaviors in wireless sensor networks is an important challenge for tasks such as monitoring applications, fault diagnosis and intrusion detection. A key problem is to define and detect those anomalies with few false alarms while preserving the limited energy in the sensor network. In this paper, using concepts from statistics, we perform an analysis of a subset of the data gathered from a real sensor network deployment at the Intel Berkeley Research Laboratory (IBRL) in the USA, and provide a formal definition for anomalies in the IBRL data. By providing a formal definition for anomalies in this publicly available data set, we aim to provide a benchmark for evaluating anomaly detection techniques. We also discuss some open problems in detecting anomalies in energy constrained wireless sensor networks.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Chandan Kumar Karmakar (Dept. of Electrical and Electronic Eng, The University of Melbourne)
Understanding Ageing Effects by Approximate Entropy Analysis of Gait Variability
Abstract: Ageing influences gait patterns which in turn affects the control mechanism of human locomotor balance. The aim of this study is to investigate the relationship between approximate entropy (ApEn) and standard deviation (SD) of a gait variable (minimum toe clearance, MTC) for young and elderly gait patterns. MTC data of 30 healthy young (HY), 27 healthy elderly (HE) and 10 falls risk (FR) elderly subjects with balance problems were analyzed. The ApEn values of MTC were significantly correlated with SD of MTC in the three groups; however, such correlations were abolished in the randomly shuffled MTC data of HE and HY group. These findings have implications of understanding ageing effect on gait variability and the likely risks of tripping falls during gait. Results are also potentially useful for the early diagnosis of common gait pathologies.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Jason Jiong Jin (Dept. of Electrical and Electronic Eng, The University of Melbourne)

Optimal Flow Control for a Multi-Service Network
Abstract: Network flow control is an important technique to allocate network resources and provide Quality of Services (QoS). Currently in the literature, the most popular strategy is Optimal Flow Control (OFC). Despite great advances of OFC, we still expose some serious limitations. For example, it is initially developed for elastic traffic only and QoS balance is not guaranteed. In order to address these issues, we propose and develop an application-oriented flow control framework for multi-service networks, which includes multitude of services with different QoS requirements. It involves an optimal flow control scheme and an efficient resource allocation architecture, consisting of both source algorithm and congestion control feedback, for future communication networks including pervasive Internet, ad-hoc wireless networks and sensor networks. Special emphasis is on supporting real-time applications, which possess non-concave utility functions.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Stuart Kininmonth (Australian Institute of Marine Science (AIMS))

Great Barrier Reef Sensor Network
Abstract: The capacity to measure the environmental fluctuations that influence the coral reefs within the Great Barrier Reef is presently limited to weather stations or isolated data loggers. While this has been useful in the past to help understand events such as coral bleaching the capacity to understand the changes in temperature, light, salinity and wave action at scales that directly effect coral and fish is now at our fingertips. Using the latest developments in sensor networks collaborative research projects have been able to develop systems that provide real time spatially rich data that can directly address the issues facing the Great Barrier Reef. Key to the success of this project has been the involvement of Melbourne University, James Coook University, University of Queensland, University of California and the Australian Institute of Marine Science. This presentation will explain and highlight the progress so far.
Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Yee Wei Law (The University of Melbourne)

MASS 2007: Lessons Learnt and Future Directions
Abstract: As part of the tradition, after attending the conference MASS 2007, I will talk about some of the interest information and results I have learnt there. These include some work done on opportunistic source coding; a new networking paradigm called store-carry-forward designed for so-called delay- tolerant networks (DTN); and a new localization technique based on graph drawing. I will also talk about the research experience Dr Koen Langendoen and Dr Paul Havinga shared with me on my trip to the Netherlands, including future research directions and implementation experiences.

23/11/2007 3pm Brown Theatre, Dept of EEE, Building 193, The University of Melbourne Daniel T.H. Lai (The

      University of Melbourne)


Application of Support Vector Machines for the Detection and Diagnosis of Gait Pathologies


back       next

top of page