Noncontact multi-modal sensor fusion for sleep stage detection

Non-contact monitoring of sleep quality and early diagnosis of sleep disorders have become hot topics in recent years. We propose a new method to extract sleep related vital signals through low-cost and non-contact multi-modal sensor fusion method with Doppler radar and microphone. Based on the PSG data of patients with sleep disorders, this work has been recognized and certified by professionals in Nanjing Integrated Traditional Chinese and Western Medicine Hospital with unique design. By comparing the sleep stage classification performance of single sensor and sensor fusion algorithm, the effectiveness of the algorithm is validated.