Nersi Nazari

On the performance of parity codes in magnetic recording systems

We investigate the performance of single and multiple parity codes in magnetic recording systems. We evaluate the codes with respect to bit error rate as well as error correction code (ECC) failure rate. While multiple parity codes outperform the single parity code in bit error rate, their performance is worse with respect to the ECC failure rate.

Analysis of error sequences for PRML and EPRML signaling performed over Lorentzian channel

It is well-known that PRML and EPRML signaling systems applied to digital magnetic recording suffer in performance due to noise correlation at the output of the channel equalizer. We analyze the error sequences for PRML and EPRML equalized Lorentzian channel in order to determine the channel performance. It is shown that an exact expression for the performance of these channels, in terms of the minimum distance error sequence, can be found. These expressions takes into account the correlation of noise due to equalization. A rigorous analysis of the error sequences is performed to compute the system performance degradations due to noise correlation. The exact expressions for the equalization filter and the output of the low pass filter both in time and frequency domains are also calculated.

Performance of recording channels employing partial response signaling

Performance of magnetic recording channels employing partial response signaling is investigated. It is shown that for the case of the Lorentzian channel, with additive white Gaussian noise, a closed form expression for the signal-to-noise ratio (SNR) can be obtained. It is demonstrated that at high densities, higher order partial response (PR) systems have less equalization loss. The off-track interference is analyzed by examining the free distance of sequences at the input to the Viterbi detector. It is shown that higher order PR systems are more sensitive to off-track noise. We use the method developed here to compare PR systems with three types of modulation coding, i.e., d=O, d=1, and trellis codes. Further analysis, based on the slope of the obtained performance curve, is used to determine the optimal linear and track densities to achieve an overall areal density given an SNR budget.