Title :
Exponential convergence of adaptive algorithms
Author :
Moustakides, George V.
Author_Institution :
Dept. of Comput. Eng. & Inf., Patras Univ., Greece
Abstract :
We introduce a novel method for analyzing a well known class of adaptive algorithms. By combining developments from the theory of Markov processes and long existing results from the theory of perturbations of linear operators we study first the behavior and convergence properties of a class of products of random matrices. This in turn allows for the analysis of the first and second order statistics of the adaptive algorithms yielding estimates for the exponential rate of convergence and the covariance matrix of the estimation error
Keywords :
Markov processes; adaptive signal processing; convergence; covariance matrices; mathematical operators; statistical analysis; Markov processes; adaptive algorithms; covariance matrix; estimation error; exponential convergence rate; first order statistics; perturbations of linear operators; random matrices; second order statistics; Adaptive algorithm; Algorithm design and analysis; Convergence; Covariance matrix; Eigenvalues and eigenfunctions; Estimation error; Informatics; Markov processes; Recursive estimation; Statistical analysis;
Conference_Titel :
Information Theory, 1998. Proceedings. 1998 IEEE International Symposium on
Conference_Location :
Cambridge, MA
Print_ISBN :
0-7803-5000-6
DOI :
10.1109/ISIT.1998.708867
