Cardiac signal enhancement using normalised variable step algorithm for remote healthcare monitoring systems

Abstract

In telecardiology, it is highly essential that good resolution signals are to be presented to the doctor for immediate diagnosis. In this aspect adaptive noise cancellers (ANCs) are promising tools to enhance cardiac signals (CS) from various artefacts. To cope with this, in the proposed paper an attempt has been made to present a new ANC using normalised variable step size least mean square (NVLMS) algorithm and its variants. In order to minimise the computational complexity, we apply sign-based and maximum normalisation treatments to the weight update recursion of NVLMS algorithm. These implementations are suitable for telecardiology applications, where large signal to noise ratios with less computational complexity are required. Finally, the developed ANCs are tested on real cardiac signals obtained from the MIT-BIH database and performance measures are compared. Simulation studies confirm that the performances of the proposed models are better than the conventional LMS-based noise cancellers.