Parallel implementation of the adaptive filter under alpha-Stable Distribution Pulse noise

Author

Zhijin, Zhao ; Baicheng, Fu

Author_Institution

Sch. of Telecommun., Hangzhou Dianzi Univ., Hangzhou

fYear

2008

fDate

7-11 June 2008

Firstpage

370

Lastpage

374

Abstract

Since the convergence speed of NLMP algorithm is relatively slow, it can not be used in some situations where the demand of real-time processing is very strict. In this paper, a parallel form of adaptive filter under alpha-Stable Distribution Pulse noise is obtained by introducing parallel FIR filter form based on polynomial decomposition. The simulation results indicate that the parallel form of NLMP algorithm can obtain the Winner solution by iterative calculation. Compared to the traditional NLMP algorithm, the parallel NLMP algorithm has faster convergence speed and higher data throughputs, and it is convenient for real-time signal processing. In addition, this algorithm is simple and feasible.

Keywords

FIR filters; adaptive filters; iterative methods; signal processing; NLMP algorithm; Winner solution; adaptive filter; alpha-stable distribution pulse noise; iterative calculation; parallel FIR filter; polynomial decomposition; real-time processing; signal processing; Adaptive algorithm; Adaptive filters; Adaptive signal processing; Background noise; Convergence; Iterative algorithms; Least squares approximation; Low-frequency noise; Signal processing algorithms; Statistical distributions; alpha-stable distribution; fractional lower order statistics; normalized least mean p-norm algorithm; parallel structure;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks and Signal Processing, 2008 International Conference on

Conference_Location

Nanjing

Print_ISBN

978-1-4244-2310-1

Electronic_ISBN

978-1-4244-2311-8

Type

conf

DOI

10.1109/ICNNSP.2008.4590374

Filename

4590374