Reduced-state sequence estimation for trellis coded modulation over ISI channels

In this paper, we adopt the combined linear-Viterbi equalizer (CLVE) technique to reduce the complexity of the Viterbi algorithm (VA) that performs maximum likelihood sequence estimation (MLSE) to trellis coded transmission over intersymbol interference (ISI) channels. This technique employs a linear filter before the VA in order to shape the original channel impulse response (CIR) to some shorter desired impulse response (DIR) which allows a lower complexity VA. Furthermore, in this paper reduced state sequence estimation for high rate concatenated systems with bandwidth efficient trellis inner codes and Reed-Solomon (RS) outer codes over ISI channels are investigated. The simulation results show that the CLVE technique can provide much better error performance compared to the previously proposed techniques with the same order of complexity. The gain obtained by using CLVE is especially noted when last taps of the channel impulse response are of substantial magnitude. The results show that substantial coding gain can be achieved by employing the RS code as outer code with only minor reduction in the overall effective information rate