Channel-optimized trellis-coded quantization for channels with memory

This paper presents the design of channel-optimized trellis-coded quantization (COTCQ) for channels with memory. The considered channel model is a binary channel with additive Markov noise. The proposed COTCQ stage exploits the channel memory by incorporating the characteristics of the correlated noise in the quantizer design. We investigate the performance of the memory-optimized COTCQ system for different bit error probabilities and noise correlation coefficients. We show that for a first-order Markov channel, COTCQ significantly outperforms channel-optimized scalar quantizer (COSQ), provided that an appropriate trellis is selected.