WSN (Wireless Sensor Network) is the technology used for sensors and communication devices to monitor data, which is characterized by low power consumption, miniaturization, and low cost. The QoS(Quality of Service) of the WSN is an important index to measure its performance, the moving characteristics of the human body and the working environment all affect the quality of the link, so that the RSSI (Received Signal Strength Indicator) is disturbed and the QoS is declined. To solve the above problems, This paper constructs an RPG-RGMM (Reference Point Group-Random Gauss-Markov Human Mobility Model) model considering the mobile characteristics of the human body, and investigates the RSSI decay pattern of ZigBee nodes moving with the body under the influence of the human body movement and the surrounding environment. The RSSI prediction value is used as an influence factor to update the selection of the relay node in the transmission management, forming an adaptive topology routing strategy. The results of the experiment demonstrate a decrease in the total packet loss rate of the adaptive topology routing strategy based on the RSSI prediction value of 10.61% compared to the traditional routing strategy. From a long-term perspective, this application has a good practical value.