Design of reconfigurable low-complexity digital hearing aid using Farrow structure based variable bandwidth filters

Abstract

A low complexity digital hearing aid is designed using a set of subband filters, for various audiograms. It is important for the device to be made of simple hardware, so that the device becomes less bulky. Hence, a low complexity design of reconfigurable filter is proposed in this paper. The tunable filter structure is designed using Farrow based variable bandwidth filter. The coefficients of the filter are expressed in canonic signed digit format. The performance can be enhanced using optimization algorithm. Here, we have explored the strength of hybrid evolutionary algorithms and compared their various combinations to select a proper coefficient representation for the Farrow based filter, which results in low complexity implementation.