Variable Step LMS Algorithm Research and Simulation
Wang Li
Abstract
Wang Li
Abstract
Although the traditional LMS algorithm has many advantages,such as simple methods and small burden in computation but for the sake of constant step,it has always contradiction in relationship between solving the steady-state error and convergence,which also makes the traditional LMS algorithm have the characteristics of slowing convergence speed.In combination of the traditional LMS adaptive filter algorithm,hereon,the step factor was improved.The variable step forward long LMS algorithm was proposed by establishing function relationship between step factor and error factor.Through MATLAB simulation and comparison of the improved algorithm with the traditional LMS algorithm,the simulation results expatiates that the improved algorithm is obviously highlighted in the convergence rate and steady-state error of the system.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Although the traditional LMS algorithm has many advantages,such as simple methods and small burden in computation but for the sake of constant step,it has always contradiction in relationship between solving the steady-state error and convergence,which also makes the traditional LMS algorithm have the characteristics of slowing convergence speed.In combination of the traditional LMS adaptive filter algorithm,hereon,the step factor was improved.The variable step forward long LMS algorithm was proposed by establishing function relationship between step factor and error factor.Through MATLAB simulation and comparison of the improved algorithm with the traditional LMS algorithm,the simulation results expatiates that the improved algorithm is obviously highlighted in the convergence rate and steady-state error of the system.
Key concepts: Convergence (economics), Computer science, Rate of convergence, Algorithm, Least mean squares filter, Adaptive filter, MATLAB, Computation