Combined concatenated coding and nonlinear equalization for magnetic recording channels

In this paper, we combine concatenated coding (with iterative decoding) and equalization for nonlinear magnetic recording channels. We modify the iterative soft-output Viterbi algorithm to incorporate a nonlinear equalizer capable of removing both precursor and postcursor (nonlinear) ISI. We adopt a simple partial erasure model against which we examine the proposed algorithm for parallel and serial concatenated codes where code rate 16/17 is considered. We demonstrate, via simulations, that improvements of about 2.5 dB at a bit-error rate of 10^-6 are possible.