The performance comparison of improved continuous mixed P-norm and other adaptive algorithms in sparse system identification

Abstract

One of the essential usages of adaptive filters is in sparse system identification on which the performance of classic adaptive filters is not acceptable. There are several algorithms that designed especially for sparse systems; we call them sparsity aware algorithms. In this paper we studied the performance of two newly presented adaptive algorithms in which P-norm constraint is considered in defining cost function. The general title of these algorithms is continuous mixed P-norm (CMPN). The performances of these algorithms are considered for the first time in sparse system identification. Also the performance of l0 norm LMS algorithm is analysed and compared with our proposed algorithms. The performance analyses are carried out through several simulation scenarios and with the steady-state and transient mean square deviation (MSD) criterion of adaptive algorithms. We hope that this work will inspire researchers to look for other advanced algorithms against systems that are sparse.