Photoplethysmogram Based Biometric Recognition for Twins

This paper presents a photoplethysmogram (PPG) based biometric recognition technique for twins. PPG devices have been widely used due to its advantages such as non-invasive, low cost and small in size which makes it a convenient analytical tool. To the best of our knowledge, little has been set pertaining to biometric recognition for twins using PPG signal. A total of six subjects from three couple of twins were used for experimentation purposes. The signals were processed using a low pass filter to remove unwanted noise. Then, multiple cycle of PPG waveforms were extracted and later, Naive Bayes (NB) and Radial Basis Function (RBF) network classifiers are used to categorize the subjects using the discriminant features. Based on the experimentation results, classification accuracies of 97% and 94% were achieved when using Naive Bayes and RBF network respectively which suggests the capability of our proposed system to identify individuals regardless whether the persons is a twin or not. The outcome also provides complimentary mechanism to detect a person besides using the current existing methods.