ACOUSTIC FEEDBACK CANCELLATION FOR DIGITAL HEARING AIDS USING SIMPLIFIED MUTLIBAND-STRUCTURED KALMAN FILTER

S.Siva Prasad* & CB Rama Rao (Author) — Mon, 11 Sep 2023 00:00:00 +0000

Acoustic coupling between the microphone and the loudspeaker is a major issue in open-fit digital hearing aids. When compared to a close-fit hearing aid, an open-fit dramatically reduces signal quality and limits the potential maximum stable gain. Adaptive feedback cancellation (AFC) is a practical method for reducing the influence of acoustic coupling. However, because to the high correlation between the loudspeaker signal and the incoming signal, it might induce bias in calculating the feedback path if not carefully considered, especially when the incoming signal is spectrally coloured, as in speech and music. For decreasing this bias, the prediction error method (PEM) is well recognized. In this paper, we proposed a simplified multi-structure Kalman filter for implementing PEM based AFC. Kalman filter allows further increase in convergence/tracking rates and the high computational complexity of generalized Kalman filter is reduced by multi-structured topology and this in turn reduce the computational complexity and also subband topology provide low processing delay. To overcome the reconvergence inability of during the change in feedback path, switched combination of SMKF and NLMS is used. Simulation results showed that the proposed algorithm performed better

A MACHINE LEARNING APPROACH FOR EARLY RENAL RISK PREDICTION USING IMPROVED SSA AND ANNGO

K. Saranyadevi* & Dr. P. Rathiga (Author) — Fri, 15 Sep 2023 00:00:00 +0000

Chronic kidney disease (CKD) is a general term encompassing a number of diverse kidney diseases with increasing frequency and prevalence as well as harmful side effects includes renal failure, cardiovascular disease, and early death. It is sometimes referred to as Chronic Renal Disease a stage of advanced renal function loss. Unfortunately, there is no cure for this ailment, but it is possible to halt its growth and mitigate the harm by diagnosing it early. As a result, it is now of utmost importance to identify such illnesses at an early stage. If the risk factors for chronic kidney disease are identified early on, timely treatment and the proper course of action may be performed. Along with mining tools, machine learning is crucial in overcoming this obstacle. In this research, we used a hybrid model and feature selection technique to construct an hybrid machine learning model to forecast CKD. With an accuracy of 99.87%, the results demonstrated that the proposed classifier performed best in the renal diagnosis method.

Computer Integrated Manufacturing Systems

ACOUSTIC FEEDBACK CANCELLATION FOR DIGITAL HEARING AIDS USING SIMPLIFIED MUTLIBAND-STRUCTURED KALMAN FILTER

A MACHINE LEARNING APPROACH FOR EARLY RENAL RISK PREDICTION USING IMPROVED SSA AND ANNGO