Enhancement in the design of Biometric Identification System based on Photoplethysmography data Girish Rao Salanke N S Research Scholar, Dept of SCSE. VIT University, Chennai Campus, Chennai, India – 600048 girishraosalanke@gmail.com Dr. Maheswari N Associate Professor, Dept of SCSE VIT University, Chennai Campus, Chennia,India-600048 maheswari.n@vit.ac.in Dr. Andrews Samraj Director, Advance Science & Technology Research Center, Salem, India - 636001 andrewsmalacca@gmail.com S.Sadhasivam Asst.Professor, Mahendra Engineering College, Tiruchengode, India. sivam.sadha@gmail.com Abstract - In the recent years, automated security systems have become one of the major concerns. Secure and reliable authentication of a person is in great demand. In this paper, we propose the applicability of Photoplethysmograph (PPG) signal for human identification. Fourier series analysis and Semi Discrete Decomposition methods are applied to extract the features that appear to offer excellent discrimination among subjects. The main obstacle while analysing a PPG signal is the presence of noise, contaminated by motion artifact. The proposed method exhibits good identification accuracies not just with the normal PPG signals, but also with the motion artifact PPG signal. Keywords - Photoplethysmograph (PPG), Motion Artifact, Fourier series Analysis, Semi Discrete Decomposition (SDD), Biometrics. I. INTRODUCTION Biometrics is a science of identifying a person using his physiological and/or behavioural characteristics[1]. Traditional biometrics like fingerprint[4], palmprint[13], face[8] and iris[5] have common weakness in their vulnerability to spoofing and even some of the other traits can be used if the person is dead. Such problems can be solved using signals like EEG[9], ECG[11] and PPG[2]. PPG signals have been used extensively in clinical diagnosis for many years. It has been recently suggested by the research communities that PPG signal can also be used as a biometric for human identification recognition. Most of the PPG biometrics work reported earlier [6][14]assumed that the PPG signal is free from motion artifact. In most environments the PPG signal is contaminated by motion artifact and surrounding light variation luring recordings. Changes in surrounding light can be rejected using the emission of modulation of signals from an infrared emitter. The most troublesome problem with PPG signal while developing an authentication system is the motion artifact. These artifacts arise mainly as a result of the air gap between the sensor and the skin which leads to poor estimation of physiological parameters from the recording. Motion artifact is mainly low frequency interference and it is random in nature. Even a slight movement by the subject while recording would then invariably disturb the contact between the sensor and the subject’s body, corresponding the PPG signal obtained during such periods to be corrupt with motion artifacts. The rest of this paper is organized as follows: Section 2 deals with theoretical framework and section 3 presents the methodology. Section 4 presents the results & discussion. Finally, Section 5 contains the conclusion. II. THEORETICAL FRAMEWORK A. Fundamentals of PPG Signal PPG signals provide a non-invasive and accurate methodology to obtain valuable physiological information such as blood oxygen saturation, heart rate and blood flow. The blood in human body is being pumped from the heart to all parts in the body by blood vessels called arteries. Blood pressure is the force of blood pushing against the walls of the arteries. Each time the heart beats it pumps out a considerable volume of blood to the arteries. Systolic pressure which is the highest blood pressure occurs when heart is in pumping motion. Diastolic pressure is lowest blood pressure when heart is in resting[7].Since blood pressures are an indirect measurement of heart beats and the blood pressure tends to change according to the time and emotion. For instance, blood pressure will rise when a subject is awaken and excited. The unit for measurement of blood pressure is in mmHg and the notation will be systolic followed by diastolic pressure. The Photoplethysmograph (PPG) signals reflect the change in blood volume caused by blood vessel expansion and Proceedings of 2013 International Conference on Green High Performance Computing March 14-15, 2013, India 978-1-4673-2594-3/13/$31.00 ©2013 IEEE ICGHPC 2013