Title :
Baseline Wander Cancellation on Trellis for Perpendicular Magnetic Recording
Author_Institution :
SK Hynix Memory Solutions Inc., San Jose, CA, USA
Abstract :
In this work, we consider a perpendicular magnetic recording channel where the data is stored perpendicular to the plane of magnetic media. The perpendicular magnetic recording provides significant aerial density gain compared to the conventional longitudinal magnetic recording. This gain comes at the cost of adding a baseline wander noise which degrades the system performance and affects the data integrity. In this article, we propose a model for the baseline wander noise. The parameters of the proposed model are computed by posing the problem as minimizing the mean square error. We demonstrate that this problem can be appropriately simplified and posed as finding a least squares problem. The proposed model on the baseline wander noise requires the knowledge of information bits to estimate the noise but this information is unavailable at the time of decoding. We propose a novel technique to cancel the baseline wander effect on the Viterbi trellis. This cancellation is done on the fly in the trellis and obviates the need of the information bits yet provides the performance gain close to the genie performance. Simulation and experimental results corroborate the analysis of the proposed algorithm and demonstrate that the proposed algorithm provides the performance gain.
Keywords :
Viterbi decoding; Viterbi detection; data integrity; least squares approximations; maximum likelihood detection; mean square error methods; perpendicular magnetic recording; trellis codes; BLW; MLSE; Viterbi detector; Viterbi trellis; aerial density gain; baseline wander cancellation; baseline wander noise; data integrity; decoding time; longitudinal magnetic recording; magnetic media; maximum likelihood sequence estimation; mean square error; perpendicular magnetic recording; trellis; Channel models; Equalizers; Maximum likelihood decoding; Noise; Perpendicular magnetic recording; Viterbi algorithm; Baseline wander (BLW); Viterbi detector; maximum likelihood sequence estimation (MLSE);
Journal_Title :
Communications, IEEE Transactions on
DOI :
10.1109/TCOMM.2015.2461672