Design and implementation of pipelined SDF FFT architecture for sustainable industrial noise suppression in Digital Hearing Aids

Abstract

This article implements Industrial Noise Reduction for Digital Hearing Aids, and the proposed solution uses the R22SDF FFT architecture to perform discrete time to frequency conversion. Acoustic feedbacks cause the Hearing Aid to develop a high-pitched tone. The 64-point Pipelined Radix 22 Single Path Delay Feedback (R22SDF) FFT Processor reduces these disturbances. In FFT, O (Nlog2N) and O (N/2log2N) minimize the product count and summation order, respectively. This paper presents a new solution based on a 64-point Pipelined R22SDF FFT architecture to reduce processing delay in existing Digital Hearing Aid processors. MATLAB Simulink and HDL Software were used to generate the simulated output. In the experimentation the hearing Aid input noises from different industrial environment is fed to proposed SDF FFT and filtered using octave filter to achieve high quality speech with SNR greater than 20 db. Using the proposed SDF FFT architecture, speech quality from hearing aid under different noise environment and signal under different environment shows 85% more clarity when compared to existing architectures.