“Context inference techniques for a wearable exercise support system” by Kawahara, Sugimoto, Arimitsu and Morandini

Computer-aided health care services have become more and more popular. This expansion has provided new means to monitor individuals’ health and to provide efficient and accurate feedback. However, it still remains difficult for people to follow fitness and/or weight loss programs because of some lack of discipline or determination. Indeed, most of these programs enforce a workout often too hard for an individual’s physical and psychological capacities. We envision that wearable computers, an integral part of our every day outfits, can assist us in dealing with such problems. In this paper, we propose the ”e-coaching” wearable computing outfit, aiming at providing interactive coaching services for efficient body regulation. Our e-coaching outfit is equipped with multiple sensors and a small PC. The coaching service processes multiple sensor readings about the user’s body and its environment. From this data, it infers a user’s activity status or context and generates coaching messages to guide the user’s training in realtime. We have implemented the e-coaching system and evaluated its context inference accuracy using resource-limited wireless sensor nodes.

1. Kawahara, Y., Hayashi, T., Tamura, H., Morikawa, H., and Aoyama, T. Privacy issues in the design of context-aware content delivery services. In Proc. of Workshop of Ubiquitous Privacy, UbiComp2004.