Optimal order of the differentiator selection in noise removal of ECG signals

Abstract

A recording of the heartbeat known as the Electrocardiogram (ECG), sometimes may get corrupt by the noise from various sources. The actual frequency of power and also its harmonics can vary based on the device and its location. A simple way to be able to eliminate the noise is the filtering of a signal using a notch filter which is based on its frequency and vicinity that may bring down the quality of an ECG since there may be components of frequency in the heartbeat as well. In order to circumvent such a loss of information, there is an optimal order for the filter that needs to be used. Fractional calculus is that branch of Mathematics that consists of the differentiation and the integrations belonging to a non-integer order. This has been migrating from the mathematicians and their theoretical realms and are also applied to several branches of engineering that may be interdisciplinary. This type of transfer functions of a Fractional Order (FO) based filters has been represented by the Fractional Order Differential (FOD) equations. The filters will then be realized by using some order elements that are fractional. For the purpose of this work, a Shuffled Frog Leaping Algorithm (SFLA), a Particle Swarm Optimization (PSO) along with a hybrid SFLA-PSO have been proposed. This proposed filter obtains input from the source of noise and the patients and the results proved that the proposed technique was able to achieve better performance than the other techniques.