Robust Adaptive Pseudo Affine Projection Algorithm for Censored Regression

Abstract

Currently censored observation systems have widely used in various business applications related to adaptive signal processing, but the traditional adaptive processing algorithm has poor convergence performance in identifying censored regression systems. The least mean square algorithm of censored regression (CR-LMS) has been proposed to solve the above problem. However, the CR-LMS algorithm will not achieve good convergence under colored inputs. Therefore, a robust pseudo affine projection algorithm based on censored regression (CR-R-PAP) is presented for the parameter estimation problem of censored regression systems under colored inputs. Compared with the pseudo affine projection algorithm based on the censored regression (CR-PAP) algorithm, the CR-R-PAP algorithm can exclude adverse effects and improve convergence performance in environments containing impulse noise. The computer simulation results show that the performance of the proposed algorithm is better than the existing algorithm in the pulse environment with different background noise.