Iterative Joint a Priori Probability for Turbo Decoding in Non-Uniform Track on Granular Media

This research presents a non-uniform track configuration in which the main track is twice as wide as its adjacent tracks, enabling high-quality soft information and enhancing adjacent-track detection via low-density parity-check (LDPC)-based turbo decoding. We then propose an iterative joint a priori probability (JAPP) detection method for turbo decoding in non-uniform track systems, in which highly-reliable information from the wider track is fed back to improve the reliability of adjacent track detection in the branch metric calculation step. Moreover, to maximize robustness against inter-track interference (ITI), the reader offset is also optimized to achieve a better bit-error rate (BER) performance. The simulation results demonstrate that non-uniform iterative JAPP configurations significantly outperform conventional uniform configurations, even at higher areal densities, achieving substantial BER improvements at an optimal reader offset of 20%.