Particle Swarm Intelligence Based Univariate Parameter Tuning of Recursive Least Square Algorithm for Optimal Heart Sound Signal Filtering

Abstract

Heart Sound Signal (HSS) is considered as one of the important bio-signals. It carries vitalinformation about the heart functions. For bio-acoustic observations, the HSS is diagnosed andrecorded with auscultatory procedures. During auscultation, the noisy components gets addedalong with the reading. The physician’s individual diagnostic experience, ecological noise andthe intersection of heart and lung sound signal (LSS) are considered as the major noisycomponents in HSS diagnosis. Suppression of LSS from the HSS is a challenging task. Due toits quasi stationary nature, adaptive filtering techniques are used for the noise removal. In thispaper, Recursive Least Square (RLS) adaptive algorithm is proposed to obtain the HSS fromthe noisy mixture. Faster convergence is a benefit in selecting RLS algorithm over otheradaptive algorithms. The forgetting factor is one of the important parameters of RLS whichdefines the convergence. The RLS performance is improved by choosing an optimal forgettingfactor. A Particle Swarm Optimization (PSO) based search algorithms are deployed foroptimization. To enhance the implementation time, a Dynamic Neighbourhood LearningParticle Swarm Optimizer (DNL-PSO) is analysed. In DNL-PSO, each particle studies from itsknowledge in dynamically varying neighbourhood that prevents early convergence. The normalHSS with different LSS interference is taken to assess the RLS filter performance. In this paper,the RLS algorithm performance is compared with Least Mean Square (LMS) adaptivealgorithms. Various metrics are used to compare the performance of both RLS and optimizationalgorithms.