Signal processing and coding techniques for perpendicular recording

Non-stationary media noise is one of the major disturbances that occur in higher density perpendicular magnetic recording channels. The strongly signal-dependent and correlated statistical nature of this noise means that it significantly worsens error-rates of traditional partial-response maximum likelihood (PRML) data-detection. As densities of recording become increasingly high, thermal stability issues place serious restrictions on the use of smaller grained recording media to reduce the media noise. To mitigate requirements for the noise reduction, we have introduced a sub-optimal ML detection scheme for media-noise-dominant channels. This scheme is based on the estimation of signal-dependent correlation-sensitive (SDCS) metric for soft-output postprocessing detection. By focusing on correcting only a few most-likely error-events in PRML detection, this media-noise soft-output detection scheme provides a good tradeoff between detection performance and computational complexity.