Frequency-Domain Functional Links For Nonlinear Feedback Cancellation In Hearing Aids

Abstract

The problem of feedback cancellation can be seen as a function approximation task, which often is nonlinear in real-world hearing assistive technologies. Nonlinear methods adopted for this task must exhibit outstanding modeling performance and reduced computational cost. To this end, in this paper, we propose a functional link adaptive filter tailored for the problem of feedback cancellation in hearing aids. The proposed model involves an adaptive exponential nonlinear expansion to better model speech signals, and a partitioned-block frequency-domain adaptation to reduce the computational cost and allow for online processing. Experimental results prove the effectiveness of the proposed method in hearing assistive problems.