Ensemble empirical mode decomposition of photoplethysmogram signals in biometric recognition

• Main Authors: Alonzo, Lea Monica B., Co, Homer S.
• Format: text
• Published: Animo Repository 2019
• Subjects: Biometric identification; Plethysmography; Machine learning; Hilbert-Huang transform; Manufacturing
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/2303; https://animorepository.dlsu.edu.ph/context/faculty_research/article/3302/type/native/viewcontent
• Institution: De La Salle University

This research focuses on using photoplethysmogram (PPG) signals for biometric recognition. Specifically, the biometric traits studied are the ensemble empirical mode decomposition (EEMD) and power spectral density (PSD) of the PPG signals. The classifiers used for testing the performance of the algorithm were K-nearest neighbors algorithm (KNN), support vector machine (SVM), and random forest (RF). Training, testing, and k-fold cross validation were done using data from public database. PPG was found to be suitable for biometric recognition, although with weakness that may be addressed through gathering and training of larger sets of data. © 2019 IEEE.