Investigating the on-board data processing for IMU-based sensors in motion tracking for rehabilitation

This paper investigates the effects of shifting computation-intensive workload for motion tracking based on sensor network from a dedicated remote server to an embedded microcontroller. The importance of battery life is exacerbated for devices designed to work in biomedical field as the prolonged operability could make a vital difference in mobile contexts. On-board processing usually drives to less communication, so the balance between the power consumed by more operations and that saved by limiting the radio transmissions has been evaluated. A measurement station has been realized to measure the energy budget needed to let more patients be monitored simultaneously and, at the same time, to increase the number of sensor nodes working together in the network. Finally, functional equivalence of the implementation has been proven.