Personalized Real-Time Monitoring of Amateur Cyclists on Low-End Devices

Abstract

Enabling real-time collection and analysis of cyclist sensor data could allow amateur cyclists to continuously monitor themselves, receive personalized feedback on their performance, and communicate with each other during cycling events. Semantic Web technologies enable intelligent consolidation of all available context and sensor data. Stream reasoning techniques allow to perform advanced processing tasks by correlating the consolidated data to enable personalized and context-aware real-time feedback. In this paper, these technologies are leveraged and evaluated to design a Proof-of-Concept application of a personalized real-time feedback platform for amateur cyclists. Real-time feedback about the user’s heart rate and heart rate training zones is given through a web application. The performance and scalability of the platform is evaluated on a Raspberry Pi. This shows the potential of the framework to be used in real-life cycling by small groups of amateur cyclists, who can only access low-end devices during events and training.