Set-membership affine projection algorithm

摘要

This letter presents a new data selective adaptive filteringnalgorithm, the set-membership affine projection (SM-AP) algorithm. Thenalgorithm generalizes the idea of the set-membership NLMS (SM-NLMS)nalgorithm to include constraint sets constructed from the past input andndesired signal pairs. The resulting algorithm can be seen as anset-membership version of the affine-projection (AP) algorithm with annoptimized step size. Also, the SM-AP algorithm does not tradenconvergence speed with misadjustment and computational complexity asnmost adaptive filtering algorithms. Simulations show the goodnperformance of the algorithm, especially for colored input signals, innterms of convergence, final misadjustment, and reduced computationalncomplexity