Lie detection is a technique of discovering if a person was telling the truth or a deceit. Therefore, to handle this matter, the field of lie detection has recently garnered significant attention. The obtained signal underwent preprocessing and was subsequently input separately into two window types of PSD (Hamming and Hanning). In both datasets, the outcome of this proposed method indicated a decrease in delta, theta, and alpha brainwave frequencies when the subject was telling untruth. This decrease in these frequency bands was attributed to relaxing. In the EXP dataset, the levels of beta and gamma were shown to be higher while the subject was lying. This finding is more realistic because beta and gamma were related to thinking and problem-solving. However, the same pattern was not observed in the Dryad dataset. The proposed technique was assessed using different window sizes, including 256, 512, and 1024. Based on the evaluation results, the EXP dataset provides a more accurate and realistic outcome. This demonstrates the quality of the data and the effectiveness of the protocol used for the experiment. The technique suggested offers a novel technique for lie detection that supports existing methods described in the literature. It is characterized by its precision, scalability, and fault tolerance.