A New Criterion of Information Theoretic Optimization and Application to Blind Channel Equalization

초록

Blind equalization techniques have been used in multipoint communication on which the research on the internet has focused. In this paper, a criterion of minimizing Euclidian Distance between two PDFs for adaptive blind qualizers has been presented. In order for ED expressed with Parzen PDFs to be minimized, we propose to use a set of randomly generated desired symbols at the receiver so that the PDF of the generated symbols matches that of the transmitted symbols. From the simulation results, the proposed method has shown superior error performance even in severe channel environments in which CMA has shown severe performance degradation. This indicates that the proposed algorithm can be considered relatively insensitive to ESR variations compared to CMA. As a field of ITL, ED minimization using Parzen PDFs has shown possibilities of being successfully applied to blind equalization.

키워드