In the process of physical fitness training, it is an important subject of scientific physical fitness training to adjust and control the physical load intensity in real-time, accurately and effectively according to the physiological load inside the human body so as to make it consistent with the predetermined goal of the training plan. Aiming at the current demand of smartphone popularization and athlete training monitoring, this paper designs an intelligent monitoring system of physical fitness based on the Internet of Things technology. By selecting such factors as vertical jump, fast leg raising, sitting forward, height, chest circumference, percentage of body constitution, YOYO intermittent endurance running and so on, using RFID technology to mark different athletes, and after using the particle swarm optimization method of BP network to establish the evaluation model of the athletes' physical condition. Through simulation, the physical condition of athletes is accurately predicted, which provides a new scientific and technological means to improve the efficiency of physical training and make physical training scientific. INDEX TERMS Physical fitness, intelligent monitoring, The Internet of Things, BP neural network, particle swarm.