Title :
On the properties of the reduction-by-composition LMS algorithm
Author :
Chen, Sau-Gee ; Kao, Yung-An ; Chen, Ching-Yeu
Author_Institution :
Dept. of Electron. Eng. & Inst. of Electron., Nat. Chiao Tung Univ., Hsinchu, Taiwan
fDate :
11/1/1999 12:00:00 AM
Abstract :
The recently proposed low-complexity reduction-by-composition least-mean-square (LMS) algorithm (RCLMS) costs only half the multiplications compared to that of the conventional direct-form LMS algorithm (DLMS). This work intends to characterize its properties and conditions for mean and mean-square convergence. Closed-form mean-square error (MSE) as a function of the LMS step-size μ and an extra compensation step-size α are derived, which are slightly larger than that of the DLMS algorithm. It is shown, when μ is small enough and α is properly chosen, the RCLMS algorithm has comparable performance to that of the DLMS algorithm. Simple working rules and ranges for α and μ to make such comparability are provided. For the algorithm to converge, a tight bound for α is also derived. The derived properties and conditions are verified by simulations
Keywords :
adaptive filters; adaptive signal processing; convergence of numerical methods; filtering theory; least mean squares methods; adaptive signal processing; closed-form mean-square error; compensation step-size; mean-square convergence; reduction-by-composition LMS algorithm; tight bound; Adaptive filters; Adaptive signal processing; Convergence; Costing; Costs; Councils; Filtering algorithms; Least squares approximation; Robustness; Signal processing algorithms;
Journal_Title :
Circuits and Systems II: Analog and Digital Signal Processing, IEEE Transactions on
