Two Dimensional Affine Projection Adaptive Filter Algorithms with Low Computational Complexity Features

Abstract-Two-dimensional (2D) adaptive filtering is a technique which is applicable to many image, and signal processing applications. This paper extends one-dimensional adaptive filter algorithms to their corresponding 2D counterparts. Based on this extension, a 2D selective partial update APA (2D-SPU-APA) and 2D partial rank affine projection (2D-PRA) are presented. Then, by the combination of these two algorithms, a new 2D adaptive filter which is called 2D selective partial rank affine projection algorithm (2DSPR-APA) is introduced. We demonstrate the good performance and applicability of our proposed algorithms by conducting several simulations in the applications of 2D system identification and noise cancellation.