Fountain Codes for the Slepian-Wolf Problem
In this paper, we describe one solution to the two-user Slepian-Wolf problem in a certain part of the achievable region using fountain codes. Symmetric case of memoryless compression of two correlated sources is considered and modeled by a BSC channel. The compression is done by two separate compressors without any exchange of information between them. The decompressor uses a Belief propagation algorithm in conjunction with the Blind Iterative Doping strategy. Simulation results indicate performance close to the Slepian-Wolf limit.
Record created on 2007-01-16, modified on 2016-08-08