Deep Neural Network Based Media Noise Predictors for Use in High-Density Magnetic Recording Turbo-Detectors
Deep Neural Network Based Media Noise Predictors for Use in High-Density Magnetic Recording Turbo-Detectors
复制标题
用于高密度磁记录涡轮探测器的基于深度神经网络的媒体噪声预测器
DOI：
10.1109/tmag.2019.2939863
复制
发表时间：
2019
影响因子：
2.1
通讯作者：
James, Ashish
中科院分区：
文献类型：
--
作者：
Sayyafan, Amirhossein;Belzer, Benjamin J.;Sivakumar, Krishnamoorthy;Shen, Jinlu;Chan, Kheong Sann;James, Ashish
本文提出了一种用于一维硬盘驱动器（HDD）磁记录的组合Bahl-Cocke-Jelinek-Raviv（BCJR）和深度神经网络（DNN）涡轮检测架构。基于颗粒翻转概率（GFP）模型的模拟HDD读数被输入到具有1-D部分响应（PR）目标的线性滤波器均衡器。均衡器输出被提供给BCJR检测器，以便最小化由于PR掩码而引起的符号间干扰（ISI）。BCJR检测器的对数似然比（LLR）输出（沿着线性均衡器输出）然后被输入到DNN检测器，DNN检测器估计依赖于信号的媒体噪声。然后以迭代方式将介质噪声估计反馈回BCJR检测器。研究了几种基于全连接（FC）和卷积神经网络（CNN）的DNN媒体噪声估计架构。对于48 nm磁道间距和11 nm位长的GFP数据，与采用1-D模式相关噪声预测（PDNP）的BCJR检测器相比，基于CNN的BCJR-DNN turbo检测器将检测器误码率（BER）降低了0.334倍，每位计算时间降低了0.731倍。所提出的BCJR-DNN turbo检测架构可以推广到二维磁记录（TDMR）。
This article presents a combined Bahl-Cocke-Jelinek-Raviv (BCJR) and deep neural network (DNN) turbo-detection architecture for 1-D hard disk drive (HDD) magnetic recording. Simulated HDD readings based on a grain flipping probabilistic (GFP) model are input to a linear filter equalizer with a 1-D partial response (PR) target. The equalizer output is provided to the BCJR detector in order to minimize the intersymbol interference (ISI) due to the PR mask. The BCJR detector's log-likelihood-ratio (LLR) outputs (along with the linear equalizer outputs) are then input to the DNN detector, which estimates the signal-dependent media noise. The media noise estimate is then fed back to the BCJR detector in an iterative manner. Several DNN media noise estimation architectures based on fully connected (FC) and convolutional neural networks (CNNs) are investigated. For GFP data at 48 nm track pitch and 11 nm bit length, the CNN-based BCJR-DNN turbo detector reduces the detector bit error rate (BER) by 0.334× and the per bit computational time by 0.731× compared to a BCJR detector that incorporates 1-D pattern-dependent noise prediction (PDNP). The proposed BCJR-DNN turbo detection architecture can be generalized for two-dimensional magnetic recording (TDMR).
基于微磁学的叠瓦磁记录简化二维写入预补偿多磁道检测分析
影响因子：
2.1
作者：
Yao Wang;B. V. Vijaya Kumar
通讯作者：
B. V. Vijaya Kumar
具有 2D 模式相关噪声预测的多轨检测
DOI：
10.1109/icc.2018.8422905
发表时间：
2018
期刊：
2018 IEEE International Conference on Communications (ICC)
影响因子：
--
作者：
Shanwei Shi;J. Barry
通讯作者：
J. Barry
二维磁记录写入过程建模及二维神经网络均衡研究
DOI：
--
发表时间：
2011
期刊：
影响因子：
--
作者：
Akihiro Fujimoto;Yusuke Hirota;Hideki Tode;Koso Murakami;河野隆二;Y.Okamoto
通讯作者：
Y.Okamoto