ABSTRACT
In this study, we propose a method to reconstruct photoplethysmogram (PPG) waveforms from other stealthily recorded physiological signals. The proposed method focuses on the frequency characteristics between two physiological signals and reconstructs the target PPG waveform using a regression model. We investigate the feasibility of the proposed method to reconstruct target PPG signals from respiratory (RSP) and PPG signals recorded at non-genuine measurement sites using the two datasets of physiological signals. The results indicate that the proposed method achieves similarities between the target PPG and reconstructed PPG signals with correlation coefficients more than 0.860.
Subject(s)
Photoplethysmography , Respiratory Rate , Photoplethysmography/methods , ElectrocardiographyABSTRACT
To develop a photoplethysmogram (PPG)-based authentication system with countermeasures, we investigate a "presentation attack" against the authentication. The attack uses the PPG for performing measurements on various sites on each subject's body. It records PPG on a nongenuine measurement site stealthily, generates a spoofing signal based on the recorded PPG, and transmits the signal to the authentication device. To investigate the feasibility of the attack, we developed a PPG-based authentication system. We recorded the PPGs of the subjects' bodies using the developed system and investigated the feasibility of attack in the experiment. The results indicated that an attack can occur with a probability of more than 80 % under ideal conditions.
Subject(s)
Biometry , Photoplethysmography , HumansABSTRACT
This research proposes a subject identification method using PPG (Photoplethysmogram) signals towards continuous authentication. The proposed method uses feature values derived from heartbeat and respiration extracted from PPG signals by means of frequency filtering and MFCC (Mel-Frequency Cepstrum Coefficients) to identify subjects. An experiment was conducted using an open dataset containing PPG signals to investigate the identification performance of the method. The feature values were extracted from the PPG signals and classifiers were generated to evaluate the performance of the method. As a result, the proposed method was found to be capable of identifying 46 people with the accuracy of 92.9 % by using feature values derived from heartbeat and respiration.