Q-State Versus FFT and WT for Stress Detection

Mental stress is a critical issue that could be defined as mental tension due to any challenging circumstances. In this paper, we are focusing on developing a stress detection model using a dry EEG headset. Our methodology uses three signal-processing techniques—Fast Fourier Transform (FFT), Q-state, and Wavelet Transform (WT)—to ensure a full diagnostic assessment. These techniques are applied to EEG data from 17 people who were both calm and stressed. The analysis focuses on EEG channels FP1, FP2, F3, and F4, which have shown significant variations between calm and stress states. Utilizing Support Vector Machine (SVM) and Random Forest (RF) machine learning algorithms, we achieved notable diagnostic accuracy. Our findings indicate that while both SVM and RF perform well under controlled conditions, the FFT and SVM combination offers a better balance between accuracy and generalization to unseen data.