Research on sleeping position recognition algorithm based on human body vibration signal

Abstract

Sleep position monitoring plays an important role in sleep health care and intervention. In medical diagnosis, sleep posture is an important index for the diagnosis and treatment of postural sleep, breathing, heart and cardiovascular diseases. The accurate identification of sleep posture is of great significance for the diagnosis of human sleep disorders. In terms of Social Nursing, the pressure of monitoring and nursing for the elderly caused by population aging is becoming more and more severe. For the elderly who have lost or partially lost their mobility, automatic sleep position monitoring will remind them when necessary or forcibly change their sleep position through technical means, which will be of great significance to improve the sleep quality. In terms of treatment intervention, according to clinical medical research, forcibly changing patients’ sleep posture through technical means will greatly improve patients’ sleep quality and sleep structure. Human body vibration signals contain rich information, such as BCG signals, respiratory signals and so on. These signals also have a close relationship with sleeping posture. Based on this relationship, this paper proposes an algorithm for sleeping position recognition. Firstly, the intelligent mattress is used to collect human body vibration signals, and the relationship between human body vibration strength and sleeping position is obtained through a series of processes such as signal processing and model training. Compared with the traditional sleeping position recognition methods such as infrared camera, this algorithm ensures the user’s privacy and reduces the influence of human body movement on the sleeping position recognition results.