Adaptive equalization of a communication channel in a non-Gaussian noise environment

The subject of adaptive filters constitutes an important part of statistical signal processing. Adaptive filters are successfully applied in such diverse fields as communications, control, radar, sonar, and biomedical engineering. In this paper we study the use of the particle filter for adaptive equalization of a linear dispersive channel that produces (unknown) distortion. The performance of the adaptive filter is compared to that of least-mean-square (LMS) and recursive-least-square (RLS) algorithms. The main advantage of the particle filter when compared to other algorithms is its robustness when dealing with non-Gaussian noise. The particle filter showed better performance in convergence speed and root-mean-square (rms) error in case of low signal-to-noise ratio.