International Journal of Current Engineering and Technology E-ISSN 2277 – 4106, P-ISSN 2347 – 5161 ©2015 INPRESSCO ® , All Rights Reserved Available at http://inpressco.com/category/ijcet Research Article 1630| International Journal of Current Engineering and Technology, Vol.5, No.3 (June 2015) Particle Swarm Optimization algorithm based Adaptive filter for Removal of Baseline Wander Noise from ECG signal Vikram B. Galphade † and P. C. Bhaskar † † Electronics Department, Department of Technology, Shivaji University, Kolhapur (MS). India Accepted 05 May 2015, Available online 11 May 2015, Vol.5, No.3 (June 2015) Abstract Electrocardiogram (ECG) is very much susceptible to the Noise and Interference, these are may be other different noise sources. Due to the different types of Noises and Interferences ECG Signal gets corrupted and hence data will be lost. Different types of Noises and Interferences like Power Line Noise (50 Hz), Base Line Wonder Noise (0.05 Hz), Muscle contraction or External High Frequency Noise etc. This Paper focus on the Base line Wonder Noise Which is having very small frequency component of 0.05 Hz to 5 Hz. This small frequency components overlap with that of the ECG Signal, due to the overlap ECG signal information gets corrupted. Hence removing the Base Line Wonder Noise we use newly developed Swarm optimization algorithm based on Adaptive Filter. Swarm Optimization Algorithm minimize the noise signal which is present in the noisy ECG signal. Keywords: Baseline Wander Noise, Adaptive Filter and Swarm Optimization algorithm. 1. Introduction 1 WHO (World Health Organization) survey: There are 80% of human deaths occurred due to the heart related diseases. Information about the cardiac status, respirational rates positions of the Heart chambers and ventricular triggering is getting from the ECG signal (Chang C-Y et al, 2010). Electrocardiogram (ECG) signal is susceptible to the interference and noise signal which are present in the surrounding environment. These external noise and interference added with ECG signal and hence necessary information is loss from ECG signal (J. Mahil et al, 2013). Fig.1 ECG signal with different wave and signals ECG signal having different waves and intervals like P, Q, R, S, T and U as shown in fig-1 above. These waves *Corresponding author Vikram B. Galphade is a Student and P. C. Bhaskar is working as Assistant Professor gives information of different positions and issues related to the heart. This information is getting from the Electrodes which are externally attached to the body. There are mainly 12 lead system is used for acquisition of ECG signal from human body. There are different noise and interference signals: Power line Noise, Baseline Wander Noise, Electromagnetic interference and Motion artifacts. These different types of interferences and noises are added in the ECG signal at the time of ECG acquisition or recording. Due to addition of interferences and noises ECG signal gets corrupted and loss clinically important data. Hence we need to remove these noises from the ECG signal for the further processing. Baseline Wander Noise having low frequency component of 0.5 Hz to 5 Hz and varying amplitude considerably affect information of ECG signal (J.M. Leski et al, 2004). This low frequency Baseline Wander Noise is overlap with ECG signal. Baseline Wander noise causes due to patients movement of hands and legs muscles or breathing. It might be induced due to misleading measurement and annotation of the signal information. Hence it is necessary to remove Baseline Wander noise from the ECG signal. Denoising of Baseline Wander Noise is done with the Adaptive filter based on Swarm algorithm. Adaptive filter provides feature of automatic adjusting coefficients. These coefficients of Adaptive filter calculated using swarm algorithm. Simulation of the model is done on the MATLAB and hardware cosimulation is done on Sparten-3a board by using