IEICE Trans - Decoding of Separately Encoded Multiple Correlated Sources Transmitted over Noisy Channels

Decoding of Separately Encoded Multiple Correlated Sources Transmitted over Noisy Channels

Kentaro KOBAYASHI  Takaya YAMAZATO  Masaaki KATAYAMA

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E92-A   No.10   pp.2402-2410
Publication Date: 2009/10/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E92.A.2402
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Information Theory and Its Applications)
Category: Coding Theory
Keyword:
correlated sources,  multiple sources,  joint decoding,  iterative decoding,  turbo codes,

Full Text: PDF>>

Summary:
We propose an iterative channel decoding scheme for two or more multiple correlated sources. The correlated sources are separately turbo encoded without knowledge of the correlation and transmitted over noisy channels. The proposed decoder exploits the correlation of the multiple sources in an iterative soft decision decoding manner for joint detection of each of the transmitted data. Simulation results show that achieved performance for the more than two sources is also close to the Shannon and Slepian-Wolf limit and large additional SNR gain is obtained in comparison with the case of two sources. We also verify through simulation that no significant penalty results from the estimation of the source correlation in the decoding process and the code with a low error floor achieves good performance for a large number of the correlated sources.