Receive Signal Strength- Based Human Activity Recognition

A novel approach based on Received Signal Strength Indicator (RSSI) used to discriminate Human Activities for biomedical applications, is presented in this paper. The proposed method use a machine-learning algorithm with discriminative features extracted from the slow fading component of the received signal. The impact of the demeaning window size used for the extraction of the slow fading is investigated and an overall accuracy of 94% is obtained with a window size of 50ms.