Title :
Signed adaptive filtering algorithms with iterate averaging
Author :
Yin, George ; Krishnamurthy, Vikram
Author_Institution :
Dept. of Math., Wayne State Univ., Detroit, MI, USA
Abstract :
This paper develops two-stage sign algorithms for adaptive filtering. The proposed algorithms are based on constructions of a sequence of estimates using large step sizes followed by iterate averaging. It is proved that the averaged sign-error algorithm converges to the minimizer with probability one. Then the asymptotic normality of a suitably scaled sequence of the estimation error is established. The asymptotic covariance is explicitly calculated and shown to be the smallest possible. Hence the asymptotic efficiency or asymptotic optimality is obtained
Keywords :
adaptive filters; convergence of numerical methods; covariance analysis; error analysis; estimation theory; iterative methods; minimisation; sequences; asymptotic covariance; asymptotic efficiency; asymptotic normality; asymptotic optimality; averaged sign-error algorithm; convergence probability; estimate sequence; estimation error scaled sequence; iterate averaging; large step sizes; minimizer; signed adaptive filtering algorithms; two-stage sign algorithms; Acceleration; Adaptive filters; Approximation algorithms; Convergence; Costs; Estimation error; Filtering algorithms; Impedance matching; Mathematics; Stochastic processes;
Conference_Titel :
Decision and Control, 2001. Proceedings of the 40th IEEE Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-7061-9
DOI :
10.1109/.2001.980640
