Multilayer Perceptron-based Detector for a Coded Two-Dimensional Magnetic Recording

Abstract

Recently, magnetic recording technology is on the verge of reaching its recording density limit. Two-dimensional (2- D) magnetic recording (TDMR) becomes a candidate for next- generation magnetic recording because of its compatibility with the current technology and powerful 2-D signal processing tools. However, the degradation in signal quality due to 2-D interference is inevitable. In this paper, we introduce the multilayer perceptron (MLP)-based detector that can produce soft-output information in the form of a log-likelihood ratio (LLR). Such a detector can also execute turbo decoding by exchanging information with a low- density parity-check (LDPC) code. The simulation result shows that the proposed system outperforms the systems that use the conventional detector based on the Viterbi algorithm.