Selective partial update normalized least mean square algorithms for distributed estimation over an adaptive incremental network

Author

Abadi, Mohammad Shams Esfand ; Danaee, Ali-Reza

Author_Institution

Fac. of Electr. & Comput. Eng., Shahid Rajaee Teacher Training Univ., Tehran, Iran

fYear

2012

fDate

15-17 May 2012

Firstpage

1329

Lastpage

1333

Abstract

Selective partial update (SPU) strategy in adaptive filter algorithms is used to reduce the computational complexity. In this paper we apply the SPU Normalized Least Mean Squares algorithms (SPU-NLMS) for distributed estimation problem based on incremental strategy in a incremental network. The distributed SPU-NLMS (dSPU-NLMS) reduces the computational complexity while it´s performance is close to the dNLMS. We demonstrate the good performance of dSPU-NLMS in both convergence speed and steady-state mean square error.

Keywords

adaptive estimation; adaptive filters; computational complexity; distributed algorithms; least mean squares methods; SPU strategy; adaptive filter algorithms; adaptive incremental network; computational complexity; distributed SPU-NLMS algorithm; distributed estimation; distributed estimation problem; incremental strategy; selective partial update normalized least mean square algorithms; steady-state mean square error; Artificial neural networks; Simulation; Selective partial update; distributed estimation; incremental network; normalized least mean squares;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical Engineering (ICEE), 2012 20th Iranian Conference on

Conference_Location

Tehran

Print_ISBN

978-1-4673-1149-6

Type

conf

DOI

10.1109/IranianCEE.2012.6292562

Filename

6292562