부산대학교 도서관

The hard disk drive industry is transitioning into array-reader-based magnetic recording (ARMR) to provide continued growth in recording density. A 2-D equalizer acts to electronically steer the array reader to provide optimal signal pickup from the track, thereby resulting in wider and deeper cross-track profile in error-rate performance. To realize this performance, the equalizer coefficients must be matched to the location of array reader, which are not known precisely due to position errors during write and read. In this paper, we present a 2-D equalization strategy that uses location diversity and pre-adaptation to mitigate the mismatch between the reader location and equalizer coefficients. The data decisions driving the various loops are generated using three 2-D equalizers (i.e., loop equalizer) that cover the range of expected offtrack. A delayed equalization strategy that permits pre-adaptation of the equalizer coefficients is used to mitigate the effect of off-track variation in the 2-D equalizer (i.e., back-end equalizer) that feeds the iterative detector. The proposed approach shows significantly improved read performance in the presence of large offtracks.