IEEE Home | Shop IEEE | Join IEEE | myIEEE | Contact IEEE | IEEEXplore
IEEE

Michael C. Gastpar


Entity Home   |   Custom Link   |   Custom Link   |   Custom Link   |   Custom Link   |   Custom Link   |   Join Entity   |   Contact Entity
Menu
»
»
»
»
»
»
»
»
»
»

Title, Abstract, and Biography

Title: "Relaying: From Statistics to Algebra"


Abstract

A major potential in large wireless networks is cooperation: A transmission from a single node is overheard not only by the intended receiver, but by all other nearby nodes; by analogy, any receiver not only captures the signal from the intended transmitter, but from all other nearby transmitters. The pessimist's perspective on these facts has shaped communication network designs of the past decades: Clever algorithms and protocols have been devised to avoid interference. Recent work, however, has revealed many scenarios under which interference turns out to be beneficial, provided it is suitably shaped. This is often referred to as physical-layer cooperation. Most of the cooperation schemes that have been proposed to date harvest the statistical dependence of the underlying signals. By contrast, in this talk, we present novel codes that permit to exploit the algebraic structure of the interference, enabling efficient and reliable computation of functions of the involved messages. Such codes will be referred to as computation codes. They are of independent interest in applications that explicitly call for computation, such as sensor networks. More generally, the computation coding perspective is used to develop a new framework for larger networks: Inside the network, judiciously chosen functions of the messages (rather than the messages themselves) are being passed around. As soon as a receiver has sufficiently many functions, it can infer the underlying message (i.e., the bits). This is reminiscent of so-called network coding, with the important difference that in the new framework, the question of which functions of the messages should be passed around is decided according to the actual interference characteristics, which can lead to significant gains. This is joint work with Bobak Nazer (UC Berkeley).


Biography

Michael Gastpar (Ph.D. EPFL, 2002, M.S. UIUC, 1999, Dipl. El-Ing, ETH, 1997) is currently an Associate Professor in the Department of Electrical Engineering and Computer Sciences at the University of California, Berkeley. He was also a student in electrical engineering and philosophy at the Universities of Edinburgh and Lausanne, and a summer researcher in the Mathematics of Communications Department at Bell Labs, Lucent Technologies. His research interests are in network information theory and related coding and signal processing techniques, with applications to sensor networks and neuroscience. He won the 2002 EPFL Best Thesis Award, an NSF CAREER award in 2004, and an Okawa Foundation Research Grant in 2008. He is currently an Associate Editor for Shannon Theory for the IEEE Transactions on Information Theory, and he will serve as Technical Program Committee Co-Chair for the 2010 International Symposium on Information Theory, Austin, TX.



Entity Home   |    IEEE Home   |    Custom Link   |    Custom Link   |    Privacy & Security   |    Terms & Conditions