| Speaker: | Dr. Kannan Ramchandran, University of California – Berkeley |
|---|---|
| Abstract: | An intriguing vision of future wireless sensor networks is that of a very large, smart collection of very small, dumb devices. What role do signalprocessing and network coding need to play towards realizing this vision?We will provide concrete examples of how randomized network coding and distributed signal processing concepts can be used to overcome cheap, unreliable, constraint-limited network nodes provided we have a large number of them. Our examples will include distributed sampling, distributed network storage, and reliable communication with cheap, untuned radios. Finally, motivated by video sensing applications involving camera phones, video surveillance networks, etc., we will describe a new distributed video coding paradigm dubbed as PRISM (Power-efficient, Robust, hIgh compression, Syndrome based Multimedia coding). Architecturally, PRISM, based on distributed compression principles from network information theory, is very complexity-flexible and has in-built robustness to packet losses. This makes it a far superior solution to existing standards like MPEG-x for video streaming over wireless networks, as we will demonstrate using Qualcomm's CDMA 2000 1X wireless networks. |
| Biography: | Kannan Ramchandran (Ph.D., 1993, Columbia University) is a Professor of EECS at UC Berkeley, where he has been since 1999. Prior to that, he was with the University of Illinois at Urbana-Champaign from 1993 to 1999, and at AT&T Bell Laboratories from 1984 to 1990. He is a Fellow of the IEEE, and his research awards include the Elaihu Jury award for best doctoral thesis in the systems area at Columbia, two Best Paper awards from the IEEE Signal Processing Society, a Hank Magnuski Scholar award for excellence in junior faculty at UIUC, and an Okawa Foundation Prize for excellence in research at Berkeley. His current research interests include wireless sensor and ad hoc networks, multimedia networking, multi-user information and communication theory, and wavelets and multi-resolution signal and image processing. |
| Presented On: | Friday, March 18, 2005 |
| Videotape: | Click to View the Streaming Quicktime Presentation This presentation requires the free Apple QuickTime Player which can be downloaded from the following link: www.apple.com/quicktime/download |
