Reliable Method to Mask Smoking Effect on Alcoholic Intake Detection Based on Photoplethysmogram Signal Analysis

摘要

This paper proposes an advanced and reliable alcoholic-intake detection method that analyzes a photoplethysmogram (PPG) signal captured from the subject. Since smoking and physical exercise as well as alcohol intake are crucial determiners of the body's biological response, we must analyze how they influence the usefulness of the PPG signal in detecting alcohol intake. In contrast to past studies, it is revealed that the respiratory signal is very useful for alcohol-intake detection regardless of smoking and physical exercise, and a more reliable alcohol intake detection approach is demonstrated. The proposed analytical methods consider not only the harmonic intensity ratio of the PPG signal, but also the oxygen-saturation level $({rm SpO}_{2})$ in the artery.