LDPC-based channel coding of correlated sources with iterative joint decoding

This paper considers low-density parity-check (LDPC) coding of correlated binary sources and a novel iterative joint channel decoding without communication of any side information. We demonstrate that depending on the extent of the source correlation, additional coding gains can be obtained. Two stages of iterative decoding are employed. During global iterations, updated estimates of the source correlation are obtained and passed on to the sum-product decoder that performs local iterations with a predefined stopping criterion and/or a maximum number of local decoding iterations. Simulation results indicate that very few global iterations (2–5) are sufficient to reap significant benefits from implicit knowledge of source correlation. Finally, we provide analytical performance bounds for our iterative joint decoder and comparisons with sample simulation results.