WIENER FILTERING AND LMS ALGORITHM OF PERIODIC NONUNIFORMLY SAMPLED STOCHASTIC SEQUENCE

A study of Wiener filtering and an LMS algorithm of a periodic nonunif ormly sampled stationary stochastic sequence is made. It is demonstrat ed that the Winer-Hopf equation of such signals should be constructed and solved on each of the M staggered ordinals, a concept of periodic time-variant weights. Also shown is that in the LMS algorithm, M weigh t vectors should be adjusted respectively according to the staggered o rdinals. Some conclusions and simulation results are given.