LEAST MEAN P-POWER ERROR CRITERION FOR ADAPTIVE FIR FILTER

An adaptive FIR filter based on the feast mean p-power error (MPE) cri terion is investigated, First, some useful properties of MPE function are studied. Three main results are as follows: 1) MPE function is a c onvex function of filter coefficients; so it has no local minima, 2) W hen input process and desired process are both Gaussian processes, the n MPE function has the same optimum solution as conventional Wiener so lution for any p. 3) When input process and desired process are non-Ga ussian processes, then MPE function may have better optimum solution t han Wiener solution, Next, a least mean p-power (LMP) error adaptive a lgorithm is derived and some application examples are presented, Conse quently, when the signal is corrupted by an impulsive noise, the adapt ive algorithm with p = 1 is preferred, Furthermore, when the signal is corrupted by noise or interference, the adaptive algorithm with prope r choice of p mag be preferred,