A NEWLY DERIVED VARIABLE DEGREE VARIABLE STEP-SIZE LMS ALGORITHM

A new least-mean-squares-(LMS-)type algorithm that employs a time-vary ing variable step size in the standard LMS weight update recursion is introduced in this paper. The work is aimed at improving the direction al estimate, in searching for the global minimum of the mean-square-er ror surface, in an effort to increase the algorithm's speed of converg ence. First-order convergence analysis of the algorithm is developed. Additionally, expressions for the ith time constant and algorithm misa djustment are introduced. Several simulation examples are presented to compare the new algorithm with the LMS and other existing variable st ep size algorithms. Comparisons illustrate the new algorithm's possess ion of better convergence properties, under stationary and non-station ary signal conditions, when compared with the other algorithms.