From channel modeling to signal processing for bit patterned media recording

2010 2010

Other formats: Order a copy

Abstract (summary)

Bit-patterned media (BPM) recording is one method proposed to overcome the density limitations imposed by the superparamagnetic effect in continuous recording media. Channel modeling, equalization, and detection aspects of BPM recording are studied in this dissertation.

In BPM recording, each bit is recorded on a single domain "island." A read channel model for BPM recording is introduced where the signal contribution from each island is evaluated. Intersymbol interference (ISI) and inter-track interference (ITI) are observed in the model due to the considered head/media geometries. The noise that arises from write/read electronics is modeled by additive white Gaussian noise (AWGN). In the model, the main component of the media noise, which is called "island jitter", is assumed to arise from the location fluctuations of islands. Island position shift in the down-track and cross-track directions is modeled with two independent Gaussian random variables. It has been shown that the jitter-induced readback voltage is non-Gaussian. Therefore, higher order approximation for the jitter-induced readback voltage is more accurate in terms of capturing the statistical properties of this noise source.

Schemes that utilize different equalization and detection methods are compared for BPM recording channels. A maximum-likelihood (ML) bit sequence detector using the Viterbi algorithm with the modified branch metric is presented for a special case of a symmetric channel response matrix. Joint-track equalization was introduced in the literature before in the context of a single interfering track. A scheme is proposed which utilizes joint-track equalization followed by a Viterbi detector for BPM recording channels. For certain recording densities, simulation results show that the performance of this scheme is comparable to that of the much more complex schemes utilizing optimal bit detection or optimal symbol sequence detection. The proposed scheme also outperforms another scheme of the same complexity introduced in the literature.

A parametric study of ITI for BPM recording channel is presented. A surprising phenomenon is observed in the performance curves of optimal bit detectors: The detector performance improved for a certain range of increasing ITI levels for channels both with and without ISI and in the absence as well as in the presence of track misregistration (TMR). For the no-ISI case, this behavior is explained by means of an exact probability of error analysis for the maximum a posteriori (MAP) bit detector, i.e. optimal bit detector. An error event analysis of a punctured ML joint-track detector is used to understand the observed effects of ITI on system performance for channels with ISI.

Indexing (details)

Electrical engineering
0544: Electrical engineering
Identifier / keyword
Applied sciences; Bit patterned media; Channel modeling; Detection; Equalization; Signal processing
From channel modeling to signal processing for bit patterned media recording
Karakulak, Seyhan
Number of pages
Publication year
Degree date
School code
DAI-B 71/04, Dissertation Abstracts International
Place of publication
Ann Arbor
Country of publication
United States
Siegel, Paul H.; Wolf, Jack K.
Committee member
Callafon, Raymond A.; Fullerton, Eric; Hodgkiss, William S.
University of California, San Diego
Electrical Engineering (Communication Theory and Systems)
University location
United States -- California
Source type
Dissertations & Theses
Document type
Dissertation/thesis number
ProQuest document ID
Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works.
Document URL
Access the complete full text

You can get the full text of this document if it is part of your institution's ProQuest subscription.

Try one of the following:

  • Connect to ProQuest through your library network and search for the document from there.
  • Request the document from your library.
  • Go to the ProQuest login page and enter a ProQuest or My Research username / password.