An Improved Variable Step Size LMS Algorithm Based on Sigmoid Function
Year: 2014
Type: article
Abstract: LMS (Least Mean Square) algorithm performance is analyzed. Then something has been done on SVSLMS that is an improved variable step size LMS algorithm based on sigmoid function. The improved algorithm has faster convergence rate, stronger channel tracking capability, better anti-noise performance and steady-state error performance compared with SVSLMS, for it makes parameters and change with channel characteristics. Theoretical analysis is the same as computer simulation results, which proves that the performance of the improved algorithm is superior to SVSLMS algorithm.
Source: Applied Mechanics and Materials
Authors: Li Liu, Qiong Wang, Hong Ping Bai
Institution: Chongqing University of Posts and Telecommunications
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W2081010624
