SHMO: A seniors health monitoring system based on energy-free sensing

The past decades have witnessed the advancement of Wireless Sensor Networks (WSNs) promoting many potential applications in the fields of smart healthcare. Through WSNs, target areas can be monitored and activities of the elderly can be recognized with a large number of deployed sensor nodes. However, previous works suffer from their coarse portability and high susceptibility to environments. In addition, the limited energy supply for sensor nodes emerges as the biggest stumbling block and such a situation is getting worse especially considering the increasing network scale. In this paper, we provide an innovative and energy-efficient system based on energy-free RFID tags to monitor the daily activities and thus determine the physical conditions of the elderly. We achieve the activity recognition by tracking the passive RFID tags attached on the elderly based on the received backscatter signals. In general, we simplify the model of seniors daily life and only consider normal move, slow move, sitting-down and fall as the basic components. DTW and SVM are utilized to discriminate them and then a healthcare assessment system can be achieved. To verify our system, extensive experiments are conducted and experiment results demonstrate that our system achieves a high recognition accuracy of various seniors daily activities and a reliable health assessment can be reached as well.