Multitask neural network based on polarity transition soft information assistance and multi-track detection method

Abstract

The invention relates to a multitask neural network based on polarity conversion soft information assistance and a multi-track detection algorithm for a heat-assisted alternating magnetic recording system. The method mainly comprises the following steps that: in a first iteration process, read-back signals of the three middle magnetic tracks are input into the multi-task neural network to obtain balanced signals and bit soft information of the three magnetic tracks, the balanced signals and the bit soft information are input into the improved BCJR detector and the LDPC decoder, and turbo balance is formed by exchanging soft information between the BCJR detector and the LDPC decoder; and, in the second iteration process, the log-likelihood ratio output by the LDPC decoder is converted into two-dimensional polarity conversion soft information, the two-dimensional polarity conversion soft information and the read-back signals of the three magnetic tracks are input into the multi-task neural network together to obtain updated equalization signals and bit soft information, and then the bit error rate is sent to an improved BCJR detector and an LDPC decoder to obtain the bit error rate of the magnetic recording system.

Description

technical field [0001] The invention relates to the technical field of signal processing in a heat-assisted alternating magnetic recording system, in particular to a multi-task neural network and multi-track detection method based on polarity transition soft information assistance. Background technique [0002] The mainstream magnetic recording technology currently used in the market is mainly based on perpendicular magnetic recording technology. "Limits further increases in its storage density. In order to further increase the storage density, the corresponding bit size needs to be reduced, and the particle volume of the magnetic recording medium also needs to be reduced to maintain a sufficient signal-to-noise ratio. But on the other hand, in order to ensure the thermal stability of magnetic recording medium particles, it is necessary to significantly increase their anisotropy, which poses a serious challenge to the write magnetic field that can be generated by the curren...

AI Technical Summary

Problems solved by technology

However, the current research on heat-assisted alternating magnetic recording systems mainly focuses on the design of new read-write systems and the optimization of read-write performance, and there are few corresponding researches and reports on signal processing algorithms.

Images