Design Of A Lie Detection System Using PPG And GSR Signals Based On Linear Methods

Lie screening Systems which are detecting the non-recognition of the truth of a person's speech suspected of lying. Characteristics of screening systems are cheaper than detecting systems, determination based on a specific factor, less accuracy, and etc. According to the results of various experiments on polygraph data, Photo Plethysmography signals and skin electrical resistance are mostly depend on mental states, especially stress. Also, the variations of these signals play a crucial role in the general conclusion of the parameters changes. In this paper twenty lie detection data recorded at the Intelligent Signal Processing Research Institute were used. After preprocessing, temporal, frequency and temporal-frequency features are extracted. Then, PCA principal component analysis was used to extract the optimal features and reduce the dimension by T-Test method. The number of optimal properties by T-Test method is eleven. Finally, the accuracy of the LDA classifier is used, is ۷۵%.

Keywords:

Lie detector systems , screening systems , photo plethysmography signal , electrical skin resistance signal , linear separator.