Variable Step Size LMS Algorithm Based on Function Control

摘要

This paper proposes a function-controlled variable step size least mean square (VSLMS) algorithm for channel estimation in low-SNR or colored input signals. The proposed method aligns the step size update with the steady-state error and alleviates the impact of high-level noise. It improves the filter performance in terms of fast convergence rate and low misadjustment error. Simulation results demonstrate the effectiveness and verify the theoretic analysis of the proposed VSLMS algorithm.