MAP decoding of MTR codes in multiple-head magnetic recording systems

This paper considers maximum a posteriori (MAP) decoding of the maximum transition run (MTR) codes, over two-head digital magnetic recording system. Decoding is based on Bayesian algorithm and computes a posteriori log-likelihood ratios (LLR) from codewords of the MTR code. In this paper we present simulation results for performance of the joint soft detection-decoding of a concatenation of MTR code and low-density parity-check (LDPC) code over the two-track two-head extended class four partial response channel (E²PR4). In channels with low inter-track interference (ITI), the joint detector-decoder exhibits 1.7dB gain.