Title :
Analysis of the data-reusing LMS algorithm
Author :
Roy, Sumit ; Shynk, John J.
Author_Institution :
Dept. of Electr. Eng., Pennsylvania Univ., Philadelphia, PA, USA
Abstract :
A variant of the popular LMS (least mean square) algorithm, termed data-reusing LMS (DR-LMS) algorithms, is analyzed. This family of algorithms is parametrized by the number of reuses (L) of the weight update per data sample, and can be considered to have intermediate properties between the LMS and the normalized LMS algorithm. Analysis and experiments indicate faster convergence at the cost of reduced stability regions and additional computational complexity that is linear in the number of reuses
Keywords :
computational complexity; convergence of numerical methods; least squares approximations; signal processing; stability; computational complexity; convergence; data-reusing LMS algorithm; least mean square; stability regions; weight update per data sample; Algorithm design and analysis; Computational complexity; Computational efficiency; Context; Convergence; Data analysis; Data communication; Least squares approximation; Signal processing algorithms; Stability analysis;
Conference_Titel :
Circuits and Systems, 1989., Proceedings of the 32nd Midwest Symposium on
Conference_Location :
Champaign, IL
DOI :
10.1109/MWSCAS.1989.102053
