An Improved Variable Step Size LMS Adaptive Filtering Algorithm
Wang Hua-kui
Abstract
Wang Hua-kui
Abstract
The conventional fixed step size LMS algorithm is difficult to result in fast convergence speed and low steady state error simultaneously.On the basis of the problem,an improved LMS adaptive algorithm is proposed in this paper,which combine variable step size algorithm and transformdomain algorithm,it can get good convergence and misadjustment property.The results of simulating manifest the performance of the new algorithm is better than others,especially in low SNR situation.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
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.
The conventional fixed step size LMS algorithm is difficult to result in fast convergence speed and low steady state error simultaneously.On the basis of the problem,an improved LMS adaptive algorithm is proposed in this paper,which combine variable step size algorithm and transformdomain algorithm,it can get good convergence and misadjustment property.The results of simulating manifest the performance of the new algorithm is better than others,especially in low SNR situation.
Key concepts: Convergence (economics), Least mean squares filter, Adaptive filter, Variable (mathematics), Algorithm, Computer science, Rate of convergence, Adaptive algorithm