Cyclist training monitoring system based on wireless sensor network

Abstract

Recent innovation of technology in wireless sensor network (WSN) has eased the deployment of WSN in many applications such as health monitoring system. This research presents a cyclist training monitoring system that is equipped with a set of sensors using the WSN technology. This enables continuous monitoring process of cyclist training that can be done anytime and anywhere. A stable and reliable wireless cyclist monitoring system with minimum data loss is vital to establish a smart and efficient sports management program that can lead to better quality outcomes of cyclist training. This cyclist training monitoring system has been developed and tested in real cyclist training environment in velodrome. The system is designed based on WSN that is linked to the cloud network on the Internet. Using TelG node as the basis, customized transceiver nodes are developed to establish the WSN. These nodes have been built with 30% reduction in size from the existing nodes. Seven measurements were conducted to investigate several factors that affect the packet loss rate before the system architecture was constructed. The factors that were taken into account during the measurements are the distance between the transmitter and the receiver, the height and angle of the receiver, the mobility of the transmitter, the transmission power of the transmitter, as well as the packet size and transmission rate. The results from the measurements correspond to the wireless communication theory. Based on the seven measurements, the system architecture was constructed. Several experiments were conducted in a real scenario in velodrome to measure the reliability of the system architecture. It was shown from the experiments that the proposed system is reliable even when the cyclist is moving at high speed which is 30km/h constantly. The packet loss in all experiments conducted is less than 2%, which does not give huge impact to the sensor data transmission. In addition, the results have shown that the proposed system can produce minimum end-to-end delay which is at 11ms when packet size is below 20 bytes which can be neglected.