Abstract
With regards to the problem that between the convergence and the steady-state error based on the existing variable step size least mean square(LMS) algorithm. So we bring up an new algorithm(LVSLMS) based on logarithmic function. First, the content and disadvantages of the conventional fixed-step LMS algorithm are introduced. In addition, some conventional variable step LMS algorithms are introduced. Second, a new variable step size LMS algorithm is raised. Comparing with [15] and [18], LVSLMS has good performance. Theoretically speaking, analysis simulation results display that LVSLMS can improve the convergence speed. Moreover, LVSLMS maintains the accuracy of the traditional algorithm. What is more, LVSLMS has made a great contribution to improving the steady-state error. So LVSLMS has strong practical value.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
