A New Variable Step Size LMS Adaptive Filtering Algorithm

• Published in: 2012 International Conference on Computer Science and Electronics Engineering Vol. 2; pp. 265 - 268
• Main Authors: Ao Wei, Xiang Wan-Qin, Zhang You-Peng, Wang Lei, Lv Chun-Ying, Wang Zheng-Hua
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.03.2012
• Subjects: Adaptive arrays; Adaptive filters; Algorithm design and analysis; Convergence; convergence rate; Least squares approximation; LMS algorithm; Signal processing algorithms; Steady-state; steady-state maladjustment; variable step size
• ISBN: 9781467306898; 1467306894
• DOI: 10.1109/ICCSEE.2012.115

Focus on the inconsistency between the convergence rate and the steady state error in LMS algorithm, this paper proposed a new variable step size LMS(VSS-LMS) algorithm, by constructing a nonlinear function between the step factor μ and the error signal e(n). The algorithm which modified the weight vector appropriately achieving the optimal by adjusting the step size parameters, could improve the convergence rate and steady-state error performance effectively. The simulation analysis showed that, this algorithm had faster convergence rate speed and smaller steady-state maladjustment compared with the basic LMS algorithm and parts of the similar variable step size algorithm, it fatherly validated this algorithm was superior to existing algorithms.