DocumentCode :
1803533
Title :
Mean-square analysis of continuous-time distributed estimation strategies
Author :
Nascimento, Vitor H. ; Sayed, Ali H.
Author_Institution :
Dept. of Electron. Syst. Eng., Univ. of Sao Paulo, Sao Paulo, Brazil
fYear :
2012
fDate :
4-7 Nov. 2012
Firstpage :
1503
Lastpage :
1507
Abstract :
In this work, we study the mean-square-error performance of a diffusion strategy for continuous-time estimation over networks. We derive differential equations that describe the evolution of the mean and correlation of the weight-error vector, and provide expressions for the steady-state mean-square deviation and excess mean-square error measures. Simulation results illustrate a good match between the theoretical model and the practical results.
Keywords :
continuous time systems; differential equations; distributed control; learning systems; mean square error methods; parameter estimation; continuous-time distributed estimation strategy; continuous-time estimation; continuous-time learning strategy; differential equation; diffusion strategy; mean evolution; mean-square analysis; mean-square error measure; mean-square-error performance; steady-state mean-square deviation; weight-error vector correlation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signals, Systems and Computers (ASILOMAR), 2012 Conference Record of the Forty Sixth Asilomar Conference on
Conference_Location :
Pacific Grove, CA
ISSN :
1058-6393
Print_ISBN :
978-1-4673-5050-1
Type :
conf
DOI :
10.1109/ACSSC.2012.6489278
Filename :
6489278
Link To Document :
بازگشت