A STUDY OF 2-D RECURSIVE LMS WITH ADAPTIVE CONVERGENCE FACTOR

적응 수렴인자를 갖는 이차원 RLMS에 관한 연구

The convergence of adaptive algorithm depends mainly on the proper choice of the design factor called the covergence factor. In the paper, an optimal convergence factor involved in TRLMS algorithm, which is used to predict the coefficients of the ARMA predictor in ADPCM is presented. It is shown that such an optimal value can be generated by system signals such that the adaptive filter becomes self optimizing in terms of the convergence factor. This algorithm is applied to real image.