Blind Equalization based on Information Theoretic Learning for Impulsive noise environments

In this paper, the development and performance study of information-theoretic learning (ITL) based blind equalizer algorithms for inter-symbol interference communication channel environments with a mixture of white Gaussian noise and impulsive noise are presented. A new developed blind equalizer algorithm based on matching two probability density functions of outputs and desired symbols employs assignment of transmitted source levels to the places of desired symbols evenly by utilizing the modulation schemes produces significant performance enhancement.. Gaussian kernel of the proposed algorithm has the effect of becoming insensitive to the large differences between impulse-infected outputs and desired level values. From the simulation results, the proposed blind algorithm shows superior performance to the CMA and/or correntropy blind algorithm in impulsive noise environments. Key-Words: Impulsive noise, blind equalization, information-theoretic learning, PDF matching, source symbol assignment, correntropy, CMA.