Performance of partial response system in perpendicular magnetic recording

Hisashi Osawa, Yoshihiro Okamoto, Yoshitake Kurihara, Yasutaka Nishida, Hiroaki Muraoka, Yoshihisa Nakamura

Research output: Contribution to journalArticlepeer-review


The error‐rate performance of the partial response maximum likelihood (PRML) system combining the partial response (PR) system with the Viterbi decoding, which is a type of maximum likelihood (ML) decoding, is investigated in perpendicular magnetic recording using a multilayered main‐pole head. Waveform approximation is carried out with regard to the isolated reproducing waveform obtained by experiment; and, based on this, the bit error rate is determined by computer simulation. First, taking into account the noise increase due to equalization, the signal‐to‐noise ratio (SNR) at the discrimination point and the bit error rates of bit‐by‐bit decoding of several PR systems using the characteristics of the multilayered main‐pole head are determined. Next, the bit error rates of the PRML system are determined for these PR systems and then compared with those for the conventional PR system. Then it is clarified that the best performance is obtained by the PR(1, 0, 0, 0, ‐1) system among these PR systems, and the improvement of SNR compared with the conventional PR system is about 1.9 dB in the case of the bit‐by‐bit decoding system and 4.1 dB in the case of the PRML system at an error rate of 10−4.

Original languageEnglish
Pages (from-to)73-81
Number of pages9
JournalElectronics and Communications in Japan (Part II: Electronics)
Issue number8
Publication statusPublished - 1994 Aug


  • Digital magnetic recording
  • PRML system
  • Viterbi decoding
  • partial response system
  • perpendicular magnetic recording

ASJC Scopus subject areas

  • Physics and Astronomy(all)
  • Computer Networks and Communications
  • Electrical and Electronic Engineering


Dive into the research topics of 'Performance of partial response system in perpendicular magnetic recording'. Together they form a unique fingerprint.

Cite this