Data storage channel equalization using neural networks

Unlike in many communication channels, the read signals in thin-film magnetic recording channels are corrupted by non-Gaussian, data-dependent noise and nonlinear distortions. In this work we use feedforward neural networks-a multilayer perceptron and its simplified variations-to equalize these signals. We demonstrate that they improve the performance of data recovery schemes in comparison with conventional equalizers. The variations of the MLP equalizer are suitable for the low complexity VLSI implementation required in data storage systems. We also present a novel training criterion designed to reduce the probability of error for the recovered digital data. The results were obtained both from experimental data and from a software recording channel simulator using thin-film disk and magnetoresistive head models.

[1]  Jaekyun Moon,et al.  Nonlinear equalization for data storage channels , 1994, Proceedings of ICC/SUPERCOMM'94 - 1994 International Conference on Communications.

[2]  G. David Forney,et al.  Maximum-likelihood sequence estimation of digital sequences in the presence of intersymbol interference , 1972, IEEE Trans. Inf. Theory.

[3]  Paul H. Siegel,et al.  Real-time recording results for a trellis-coded partial response (TCPR) system , 1993 .

[4]  Sammy Siu,et al.  Multilayer perceptron structures applied to adaptive equalisers for data communications , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[5]  A. M. Patel.  A new digital signal processing channel for data storage products , 1991 .

[6]  Dean Palmer,et al.  Identification of nonlinear write effects using pseudorandom sequences , 1987 .

[7]  S. Siu,et al.  Decision feedback equalisation using neural network structures and performance comparison with standard architecture , 1990 .

[8]  Francesco Piazza,et al.  Fast neural networks without multipliers , 1993, IEEE Trans. Neural Networks.

[9]  Hisashi Kobayashi,et al.  Application of partial-response channel coding to magnetic recording systems , 1970 .

[10]  P.H. Siegel,et al.  Modulation and coding for information storage , 1991, IEEE Communications Magazine.

[11]  Jaekyun Moon,et al.  Hardware-efficient nonlinear equalizer and its performance , 1995, Proceedings IEEE International Conference on Communications ICC '95.

[12]  G. Jacoby.  A new look-ahead code for increased data density , 1977 .

[13]  Jaekyun Moon,et al.  Performance comparison of detection methods in magnetic recording , 1990 .

[14]  Roger W. Wood,et al.  Viterbi Detection of Class IV Partial Response on a Magnetic Recording Channel , 1986, IEEE Trans. Commun..

[15]  Jaekyun Moon,et al.  Improved equalization for digital recording using nonlinear filtering and error confinement , 1994 .

[16]  Roger Wood.  Turbo-PRML: a compromise EPRML detector , 1993 .

[17]  Jan W. M. Bergmans.  Density improvements in digital magnetic recording by decision feedback equalization , 1986 .

[18]  R. Baugh,et al.  Measurement of noise in magnetic media , 1983 .

[19]  P. A. Ziperovich,et al.  Performance degradation of PRML channels due to nonlinear distortions , 1991 .

[20]  D.R. Hush,et al.  Progress in supervised neural networks , 1993, IEEE Signal Processing Magazine.

[21]  Bum S. So,et al.  Performance data for a six-sample look-ahead , 1993 .

[22]  Bing J. Sheu,et al.  A neural-based digital communication receiver for inter-symbol interference and white Gaussian noise channels , 1992, [Proceedings] 1992 IEEE International Symposium on Circuits and Systems.

[23]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[24]  Paul H. Siegel,et al.  Trellis -coded Partial Response (TCPR): An Improved Recording Method Over Prml , 1993 .

[25]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[26]  M. K. Bhattacharyya,et al.  Studies of nonlinear bit shift and partial erasure using pseudo-random sequences , 1993 .

[27]  H. N. Bertram,et al.  Transition noise model for longitudinal thin-film media , 1987 .

[28]  G. H. Lin,et al.  Experimental studies of nonlinearities in high density disk recording , 1992 .

[29]  G. H. Lin,et al.  Noise correlations in dibit recording , 1994 .

[30]  Gustavo de Veciana,et al.  Neural net-based continuous phase modulation receivers , 1992, IEEE Trans. Commun..

[31]  Roy D. Cideciyan,et al.  A PRML System for Digital Magnetic Recording , 1992, IEEE J. Sel. Areas Commun..

[32]  Jaekyun Moon,et al.  Discrete-time modeling of transition noise dominant channels and study of detection performance , 1991 .

[33]  C. Melas,et al.  Nonlinear superposition in saturation recording of disk media , 1987 .

[35]  John M. Cioffi,et al.  Performance of digital magnetic recording with equalization and offtrack interference , 1991 .

[36]  M. K. Loze.  Modelling the asymmetry of the recorded transition zone in digital magnetic recording , 1994 .

[37]  S. Haykin,et al.  Adaptive Filter Theory , 1986 .

[38]  K. Chopra,et al.  The effects of MR induced non-linearities on channel performance , 1994 .

[39]  John M. Cioffi,et al.  An adaptive RAM-DFE for storage channels , 1991, IEEE Trans. Commun..

[40]  J. Cioffi,et al.  A nonlinear model for thin film disk recording systems , 1993 .

[41]  Noise distribution measurements of recorded transitions in film media , 1993 .

[42]  G. J. Kerwin,et al.  Performance Evaluation Of The Disk Drive Industry's Second-generation PRML Data Channel , 1993 .

[43]  C.M. Melas,et al.  Adaptive equalization in magnetic-disk storage channels , 1990, IEEE Communications Magazine.

[44]  Jaekyun Moon,et al.  Simplified nonlinear equalizers , 1995 .

[45]  R. L. Galbraith,et al.  Performance evaluation of the disk drive industry's second-generation partial response maximum likelihood data channel , 1993 .

[46]  R. Wood.  Enhanced decision feedback equalization , 1990, International Conference on Magnetics.

[47]  T. D. Howell,et al.  Error rate performance of experimental gigabit per square inch recording components , 1990, International Conference on Magnetics.

[48]  H. Thapar,et al.  A class of partial response systems for increasing storage density in magnetic recording , 1987 .

[49]  J.M. Cioffi,et al.  Spectral shaping for peak detection equalization , 1990, International Conference on Magnetics.

[50]  Jaekyun Moon,et al.  Density dependence of noise in thin metallic longitudinal media , 1987 .

[51]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[52]  David G. Messerschmitt,et al.  DIGITAL COMMUNICATIONS , 2013 .