Diffusion Least-Mean Squares Over Adaptive Networks

Author

Lopes, Cassio G. ; Sayed, Ali H.

Author_Institution

Dept. of Electr. Eng., California Univ., Los Angeles, CA, USA

Volume

3

fYear

2007

fDate

15-20 April 2007

Abstract

Distributed adaptive algorithms are proposed to address the problem of estimation in distributed networks. We extend recent work by relying on static and adaptive diffusion strategies. The resulting adaptive networks are robust to node and link failures and present a substantial improvement over the non-cooperative case asserting that cooperation improves estimation performance. The distributed algorithms are peer-to-peer implementations suitable for networks with general topologies.

Keywords

adaptive estimation; adaptive signal processing; least mean squares methods; adaptive diffusion strategies; adaptive networks; diffusion least-mean squares; distributed adaptive algorithms; link failures; peer-to-peer implementations; performance estimation; Adaptive algorithm; Adaptive estimation; Adaptive systems; Aggregates; Distributed algorithms; Network topology; Peer to peer computing; Protocols; Robustness; Signal processing algorithms; Adaptive filters; Adaptive signal processing; Adative estimation; Cooperative systems; Distributed estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference on

Conference_Location

Honolulu, HI

ISSN

1520-6149

Print_ISBN

1-4244-0727-3

Type

conf

DOI

10.1109/ICASSP.2007.366830

Filename

4217860