New Adaptive Equalization Method for Partial Response Maximum Likelihood System Optimizing Error Rate Performance

Abstract

Adaptive equalization is essential in achieving wider system margins, such as tilt margin and defocus margin, in optical disk systems using partial response maximum likelihood (PRML) detection method. The most conventional method is the combination of the use of the least-mean-squares (LMS) algorithm and partial response (PR) equalization, but it does not optimize the characteristics of equalizers in terms of bit error rate (BER) performance. Integrating the concept of sequenced amplitude margin (SAM) with LMS, we developed a new adaptive equalization method, called least-mean-squared SAM error (LMSAM), which can minimize BER. After explaining the basic idea of LMSAM, we theoretically derived its implementable algorithm. We also reported favorable experimental results on the remarkable improvement of the tangential tilt margin and defocus margin with LMSAM. Furthermore, we analyzed the result from the viewpoint of frequency response and clarified why LMSAM can improve tilt margin.