Intelligent Mattress Monitoring System Based on RFID RSSI Difference Array

The nursing care of bed-ridden patients is an overworked task, in which the attitude testing of bed-assistants is an important technique. At present, there are three main approaches to detect a person's posture on the bed, i.e., pressure sensor arrays, wearable devices, and image recognition methods. While, the pressure sensor method is faced with the lack of precision and high cost of the challenge, the wearable devices face the problem of comfort, and the methods of image recognition have a risk of violating privacy. In this paper, a new method based on passive multi-label RFID array mattress system is proposed to detect the pose of a human body. This proposed methods uses the difference matrix formed by the influence of the passive label's RSSI value and the dielectric constant of human body. Through the random forest classification method, the final accuracy rate reached 96.14%. Compared with the previous works, the accuracy rate of the proposed method gets increased by 5.36% and 4.54% respectively. In addition, the system features low cost and portable layout.