Anti-Noise Based on Improved Variable Step size LMS Adaptive Filtering Algorithm

摘要

The question of how to suppress noise when there is no correlated noise source is researched The paper utilizes the difference between a narrow-band signal’s predictability and the noise’s predictability to suppress noise. The principle of anti-noise is analyzed and the weight-update formula of the variable step size LMS (least mean squares) adapive filtering algorithm is improved. The simulation results show that the improved variable step size LMS algorithm has smaller steady-state offset noise and faster convergence rate than conventional variable step size LMS algorithm.