2011 Poster Sessions : Minimum-Context Channel Decoding

Student Name : Vinith Misra
Advisor : Tsachy Weissman
Research Areas: Information Systems
Shannon's channel coding theorem demonstrates that reliable communication is possible at any rate below the capacity of a discrete memoryless channel. Implicit in the problem statement, however, is that a significant amount of contextual information may be shared between the encoder and decoder. Specifically, the decoder is assumed to know the communication rate, the blocklength, the channel statistics, a list of codewords, and the source-symbol-to-codeword mapping. How much of this contextual information is necessary, and how much can be discarded without affecting capacity? We answer this question by constructing a minimum-context decoder and demonstrating its asymptotic reliability at all rates under capacity.

Vinith Misra received the S.B. and M.Eng. degrees in electrical engineering from the Massachusetts Institute of Technology in 2008, where his thesis was awarded the David Adler Memorial M.Eng. thesis prize. He is currently pursuing a Ph.D. in Stanford University's department of electrical engineering under the support of the Stanford Graduate Fellowship and the National Defense Science and Engineering Graduate Fellowship. His research interests include information theory, signal processing, and mixed-signal circuit design, with applications to both communications and medical devices.