Fast NLMF-type algorithms for adaptive sparse system identifications

Adaptive sparse system identification (ASIDE) techniques have been successfully applied in many applications, such as sparse channel estimation and radar target detection. Normalized least mean fourth (NLMF)-type algorithms are considered as one of the stable ASIDE techniques even at low signal-to-noise ratio (SNR). However, the convergence capability of sparse NLMF algorithms is severely decreased by initial mean square error (MSE) and input variance in the high SNR regimes. To improve the convergence speed of the sparse NLMF algorithms in all SNR regions, in this paper, we propose a kind of non-constraint fast sparse NLMF-type algorithms for applying in ASIDE. Unlike the conventional methods, the proposed algorithms provides an alternative way to get rid of the restriction of SNR-dependent initial MSE and input variance. The proposed fast sparse NLMF-type algorithms are validated via computer simulations.