Robust adaptive filtering algorithms for alpha-stable random processes

A new class of algorithms based on the fractional lower order statistics is proposed for finite-impulse response adaptive filtering in the presence of alpha-stable processes, It is shown that the normalized least mean p-norm (NLMP) and Douglas' family of normalized least mean square algorithms are s pecial cases of the proposed class of algorithms. A convergence proof for t he new algorithm is given by showing that it performs a descent-type update of the NLMP cost function. Simulation studies indicate that the proposed a lgorithms provide superior performance in impulsive noise environments comp ared to the existing approaches.