Regressive channel coding with sequential decoding for embedded source coders

An adaptive error protection scheme for embedded source coders is presented. Convolutional codes of very high memory and regressive redundancy are applied to encode the data frame of the source. The channel decoder with scalable complexity and delay employs modified sequential decoding. In contrast to conventional coding systems, the principal idea of this new algorithm, the `far end error decoder (FEED)´ is to delay the first error as far to the end of the frame as possible rather than to aim at a low average error rate. The proposed system is particularly appropriate for transmitting over strongly varying channels