Reduced Complexity Neural Network Equalizers for Two-dimensional Magnetic Recording

TL;DR

This paper develops reduced complexity neural network equalizers for two-dimensional magnetic recording, achieving significant BER improvements with lower computational costs compared to traditional neural network approaches.

Contribution

It introduces reduced complexity MLP equalizers that maintain high performance while significantly lowering computational complexity for TDMR channels.

Findings

RC-MLP achieves 8.23% BER reduction over linear equalizer.

Proposed RC-MLP has 1.59 times the complexity of linear equalizer.

MLP achieves 10.91% BER reduction but with 6.6 times complexity.

Abstract

This paper investigates reduced complexity neural network (NN) based architectures for equalization over the two-dimension magnetic recording (TDMR) digital communication channel for data storage. We use realistic waveforms measured from a hard disk drive (HDD) with TDMR technology. We show that the multilayer perceptron (MLP) non-linear equalizer achieves a $10.91\%$ reduction in bit error rate (BER) over the linear equalizer with cross-entropy-based optimization. However, the MLP equalizer's complexity is $6.6$ times the linear equalizer's complexity. Thus, we propose reduced complexity MLP (RC-MLP) equalizers. Each RC-MLP variant consists of finite-impulse response filters, a non-linear activation, and a hidden delay line. A proposed RC-MLP variant entails only $1.59$ times the linear equalizer's complexity while achieving a $8.23\%$ reduction in BER over the linear equalizer.