Classification of human limb rehabilitation activities

A feasibility study on using the received signal strength measured by small wireless transceivers in order to classify typical human body rehabilitation movements is presented. Wearable wireless low-cost commercial transceivers operating at 2.4 GHz incorporate a Received Signal Strength Indicator (RSSI): the key idea is to collect the RSSI values measured between a set of wireless devices strategically placed on a human body, specifically on upper and lower limbs, to monitor some typical rehabilitation activities. The collected RSSI data are processed using Support Vector Machine (SVM) and K-Nearest Neighbor (K-NN) algorithms, in order to classify the rehabilitation activities.