Performances of sampled amplitude sequence detectors on a partial erasure magnetic recording channel

The performance of sampled amplitude sequence detectors are evaluated for a model of the magnetic recording channel. In particular, the performance of detectors currently used with run-length-limited codes having d=0 and d=1 constraints are compared including code rate loss, practical equalization and partial erasure effects. It is shown that a d=0 PR4 detector performs slightly better than a d=1 EEPR4 detector at reasonable user densities when the channel is linear. However, in the partial erasure channel model used, the d=1 detector has a significant advantage.