1 Introduction
Health monitoring systems are widely used in the filed in medicine and nursing. Usually, conventional health monitoring system constrains patients, that is, electrocardiographs and electroencephalographs constrain patients. A nonconstraining health monitoring system is essential for sleeping patients because we do not have to concern ourselves with the input sensor being detached. A number of health monitoring sensors for humans in bed have been developed [1]–[4]. In our study, we attempt to detect three biological information. First one is recognition for humans in bed. We will recognize two conditions; an empty bed or a sleeping patient in bed. Second one is to monitor heart rate. The heart beat mainly shows an automatic nervous system activity, and its rate is an important parameter to influence a blood pressure change, an aspiration system, and a temperature control and so on. The third one is to detect the behavior before getting out of bed of patients to prevent accident fall from bed.