Channel equalization for coded signals in hostile environments

In this paper the detection of Trellis Coded Modulated signals corrupted by Intersymbol Interference, Co-Channel Interference and nonlinear impairments is treated as a classification task by means of a Clustering Based Sequence Equalizer-Decoder. The receiver performs jointly decoding and equalization of trellis encoded signals. No specific model is required for the channel or for the interference and the noise, and no code knowledge is needed at the receiver. Complexity reduction of the equalizer is obtained through two sub-optimal techniques, a) Cluster's grouping and b) the M-Algorithm. The robust performance of the proposed scheme is illustrated by simulations.