To overcome the limitations of traditional physiological signal sensors in the field of emotion recognition

a non-contact emotion recognition method based on a frequency modulated continuous wave(FMCW)radar is proposed.Firstly

a non-contact physiological signal detection algorithm is designed based on the improved complete ensemble empirical mode decomposition with adaptive noise(ICEEMDAN)

which separates and reconstructs the respiratory and heartbeat signals.Subsequently

feature extraction

fusion

and optimization are performed on these two signals

and input to a support vector machine(SVM)classifier for emotion recognition.The results of emotion induction experiment based on audiovisual stimulation show that the proposed method achieves 79.62% accuracy in recognizing four emotions of happiness

relaxation

fear

and sadness.Experiments show that using non-contact physiological signal detection can effectively evaluate individual emotional state.