Enhancing detection sensitivity of alcohol intake by dividing photoplethysmogram signals

This paper uses the FFT technique to analyze the characteristics of the pulse-modulated photoplethysmogram (PPG) signals of pre- and post-alcohol-intake subjects in terms of spectral amplitude. The input PPG signal is automatically divided into short signal segments with time stamps. This paper investigates which part of the PPG signal reflects cardiac motion and so is eventually influenced by alcohol intake. The impact of alcohol intake on a specific spectrum of the PPG signal is also considered. Experiments strongly suggest that the spectral amplitude of the signal segments that reflect diastolic motion (of the order of hertz) should be applied to detect alcoholic intake.