Abstract: Environmental noise impacts the quality of life and psychological well-being of people with hearing sensitivity or autism spectrum disorder due to the lack of effective noise control strategies. People with this disorder have auditory hypersensitivity and, in many cases, hyperacusis, which increases their vulnerability to loud noises. Previous research describes the use of active noise cancellation algorithms or noise regulatory measures to protect the health and quality of life of people with the disorder. Therefore, we propose the implementation of an active domestic noise cancellation technique based on the adaptive FxLMS algorithm and implemented in in-ear headphones, aimed at reducing sounds such as fireworks, blenders, door slamming, and barking, to protect the hearing health of people with the syndrome. The project consisted of modeling the Primary Path and Secondary Path propagation routes from a database and adapting the system to the Bose QC20 in-ear headphones. The results demonstrated that the adaptive FxLMS algorithm managed to attenuate domestic noise in low and medium frequencies (<4000 Hz), confirming its adaptation capacity, although with critical dependence on the convergence coefficient and limitations associated with the type of noise and database used.

Keywords: Hypersensitivity, Hyperacusis, Adaptive algorithm, Least Mean Squares, Adaptive filtering, FxLMS, noise cancellation

Pages: 225-235