| August 30, 2010 | ||
| 12:30 pm | to | 2:00 pm |
The IEEE Canadian Atlantic Section Signal Processing and Microwave Theory and Techniques Chapter presents an IEEE Distinguished Lecture by Professor Vikram Krishnamurthy of the University of British Columbia. The lecture is open to public and the details of the lecture are as follows:
| Title: | Adaptive Filtering Games for Designing Reconfigurable Sensor Networks | |
| Speaker: | Dr. Vikram Krishnamurthy, Canada Research Chair in Signal Processing, University of British Columbia | |
| Time: | 12:30pm-2:00pm, Monday, August 30, 2010 | |
| Place: | Room B310, Sexton Campus, Dalhousie University, Halifax, Nova Scotia, Canada | |
| Refreshments: | Drinks and foods will be provided to the attendees | |
| Local Contact: | Dr. Zhizhang (David) Chen, Dalhousie University, Tel: (902) 494-6042, E-mail: zz.chen@ieee.org |
Abstract:
Decentralized awareness in a sensor network requires decentralized information processing. The idea is that if each sensor or small group of sensors can appropriately adapt their behavior to locally observed conditions, they can quickly self-organize into a functioning network, eliminating the need for difficult and costly centralized control. This talk deals with decentralized information processing and social learning in sensor networks using game theoretic methods. The talk comprises of three parts. In the first part, we describe how social learning leads to the remarkable behavior of rational herding, where all sensors eventually end up taking the same action. In the second part of the talk, we illustrate how the theory of global games gives a powerful method for designing decentralized data-aware sensor activation algorithms in dense sensor networks. We show that the Nash equilibrium of the sensor network has a simple threshold structure and exhibits a remarkable phase transition as more data is collected. In the third part of the talk we describe how decentralized adaptive filtering algorithms with regret matching can be deployed in sensor networks to guide network behavior to a correlated equilibrium. A major theme of the talk is how simple local behavior can result in sophisticated global behavior.
About the Speaker:
Vikram Krishnamurthy (F) currently holds the Canada Research Chair in Signal Processing at the University of British Columbia. Prior to 2002, he was a Chaired Professor, University of Melbourne, Australia where has served as Deputy Head of Department. He has made several contributions to the theory of bayesian estimation, stochastic sensor scheduling, and hidden markov models.
Dr. Krishnamurthy’s current research interests include computational game theory, stochastic dynamical systems for modeling of biological ion channels and stochastic optimization and sensor scheduling. Much of his recent research deals with sensor-adaptive signal processing – that is, how networked sensors can dynamically adapt their behavior to optimize the statistical signal processing. Such problems use game theory and stochastic control together with statistical signal processing.
Dr Krishnamurthy has published over 30 book chapters and 125 peer reviewed journal papers. He has served as Associate Editor, IEEE Transactions Signal Processing (2000-2005); IEEE Transactions Automatic Control; IEEE Transactions Aerospace & Electronic Systems; IEEE Transactions Circuit and Systems II; IEEE Transactions Nanobioscience; EURASIP Journal of Applied Signal Processing; and Systems & Control Letters. Dr. Krishnamurthy has received many awards for his research including the Canada Research Chair, and Queen Elizabeth II Fellowship. He is a Fellow of the IEEE and a Member, IEEE Signal Processing Theory and Methods Technical Committee (2005-present).
