Bi-Directional Pattern-Dependent Noise Prediction for Heat-Assisted Magnetic Recording With High Jitter Noise

In this paper, we first report the results of an investigation of perpendicular heat-assisted magnetic recording (HAMR) channels containing high jitter noise that is likely to be the case at very high areal densities (e.g., 4 Tb/in²). To model the HAMR channel, we use the thermal Williams-Comstock model and the microtrack model to derive the transition response, without large thermal spot approximation. Further, we propose a novel bi-directional pattern-dependent noise prediction (biPDNP) detector to improve the performance of the HAMR channel under high jitter noise conditions. The biPDNP detector utilizes backward linear prediction in the noise prediction process, as well as the conventional forward linear prediction. At bit error rate of 10^-3, biPDNP detector offers 0.6-1.4 dB performance gain over the conventional PDNP detector.