Iterative Joint Source-Channel Decoding for H.264 video transmission using virtual checking method at source decoder

This paper proposes an Iterative Joint Source-Channel Decoding (IJSCD) scheme for error resilient transmission of H.264 compressed video over noisy channels. Here both the redundancy systematically added by the channel coding and the source semantic residual redundancy in the compressed video sequence are exploited, in a co-operative manner, to detect and correct transmission errors. Specifically, the soft values produced by the maximum a posteriori (MAP) decoder are used to generate a list of slice candidates for each slice, which are then verified for semantic errors. A new semantic checking method is proposed to accelerate the semantic verification process.