DefinePK

Abstract

This study explicates the practicability of using acceleration plethysmogram (APG) signal in biometric identification. The introduction of APG signal is initiated from the congenital of photo plethysmogram (PPG) signal since APG signal has been widely known as the second derivative of PPG signal. Previous researchers claimed that APG signal elucidates more information as compared to PPG signal. For this reason, the robustness and reliability of APG signal as biometric recognition is demonstrated. A total of 10 subjects obtained from MIMIC II WAFEFORM Database (MIMIC2WDB) which provides PPG signals with a 125 Hz sampling frequency are used as test samples. The signals are then differentiated twice to obtain the APG signals. Then, discriminative features are extracted from the APG morphology. Finally, these APG samples were classified using commonly known classification techniques to identify individuals. Based on the experimentation results, APG signal when using Multilayer Perceptron gives an identification rate of 98% as compared to PPG signal of 76% for the same waveform. This outcome suggests the feasibility and robustness of APG signals as a biometric modality as an alternative to current techniques.