Body sensor networks consist of a group of wireless sensors with various monitoring capacities. The system transmits data that BSNs collect for analysis by, for example, doctors. In a BSN activity-recognition system, sensor sampling and communications quickly deplete battery reserves. Reducing sampling and communication saves energy, but usually at the cost of reduced recognition accuracy.
At the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS), researchers from the College of William and Mary presented a paper proposing AdaSense, a framework that reduces the BSN sensors sampling rate while meeting a user-specified accuracy requirement. AdaSense utilizes a classifier set to perform either multiactivity classification that requires a high sampling rate or single-activity event detection that demands a very low sampling rate. Furthermore, it uses a novel genetic-programming algorithm to determine optimal sampling rates.
“AdaSense: Adapting Sampling Rates for Activity Recognition in Body Sensor Networks” and other papers from ICNC 2013 are available to both IEEE Computer Society members and paid subscribers via the Computer Society Digital Library.