Turbo-detection for Multilayer Magnetic Recording Using Deep Neural Network-based Equalizer and Media Noise Predictor

In the hard disk drive (HDD) industry, new technologies are being developed to increase density such as two-dimensional magnetic recording (TDMR). TDMR utilizes 2D signal processing without changes to existing magnetic media to get remarkable density gains [1] . In multilayer magnetic recording (MLMR), an additional magnetic media layer is vertically stacked to a TDMR system to achieve additional density gains [2] , [3] . We study deep neural network (DNN) based methods for equalization and detection for MLMR, using a realistic grain switching probability (GSP) model [4] for generating waveforms.