Robust feedback cancellation in hearing aids in the presence of outliers

Author

Vasundhara ; Panda, Ganapati ; Puhan, N.B.

Author_Institution

Sch. of Electr. Sci., Indian Inst. of Technol. Bhubaneswar, Bhubaneswar, India

fYear

2014

fDate

20-21 Feb. 2014

Firstpage

25

Lastpage

29

Abstract

The feedback cancellation scheme in hearing aids mainly employs Filtered-XLMS (FXLMS) algorithm However, the conventional FXLMS algorithm doesnot provide satisfactory performance in the presence of outliers acquired by the microphone. In this paper, Filtered-X Wilcoxon LMS (FXWLMS) and Filtered-X least mean log square (FXLMLS) are being introduced as two robust learning algorithms for effective feedback cancellation in the presence of outliers. The simulation studies demonstrate the superior feedback cancellation performance of FXLMLS and FXWLMS over FXLMS in the presence of outliers.

Keywords

adaptive filters; feedback; hearing aids; learning (artificial intelligence); least mean squares methods; medical signal processing; FXLMLS; FXLMS algorithm; FXWLMS; feedback cancellation scheme; filtered-X Wilcoxon LMS; filtered-X least mean log square; filtered-XLMS algorithm; hearing aids; microphone; outliers; robust learning algorithms; Auditory system; Filtering algorithms; Hearing aids; Microphones; Robustness; Signal processing algorithms; White noise; Adaptive filters; Wilcoxon norm; hearing aids; identification; least mean log squared;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Integrated Networks (SPIN), 2014 International Conference on

Conference_Location

Noida

Print_ISBN

978-1-4799-2865-1

Type

conf

DOI

10.1109/SPIN.2014.6776915

Filename

6776915