With the development of low-power wide area networking technology and micro sensors, they have been widely used in wearable devices, achieving better transmission of vital sign monitoring data. However, existing technologies such as Bluetooth and Wi Fi have difficulties in networking in scenarios such as nursing homes and other multi node network monitoring, and due to sensor volume limitations, blood pressure cannot be effectively monitored. This article designs and implements an ad hoc network human vital sign detection system based on LoRa communication technology. After the system node solves the heart rate, blood oxygen, body temperature and human PPG waveform feature vector through local calculation, a self-organizing network is constructed using the LoRa RF chip to upload multi node data to the gateway and then to the server's network. The server uses BP neural network to calculate blood pressure data, and the analysis results are synchronized to the node display through the monitoring network. The experimental results show that the system can achieve normal communication among 7 node terminals, with a communication distance of up to 1.55km, and physiological parameter errors within 5%. The blood pressure monitoring effect meets the American Association for the Promotion of Medical Devices (AAMI) blood pressure measurement standard. The system has broad application prospects in nursing, elderly care monitoring and other scenarios.