Indoor Device-Free Free Activity Recognition Based on Radio Signal
Abstract: Activity recognition based on a wireless signal has received a lot of attention recently. Only requiring some modification on Access Point (AP), this type of approach is applicable to indoor environments. Many approaches require the line-of-sight sight signal, while activity recognition based on a wireless signal in nonnon line-of-sight sight (NLOS) environment remains a challenging problem. More importantly, how to distinguish the velocity of humans humans is still an open problem. In this paper, we present an approach for device-free device free movement recognition that leverages the radio frequency signal in the NLOS environment. We extract features from the received packet to classify different moving directions and velocities. We propose to distinguish human height based on the frequency feature of the received packet as well. Our experiment shows that the proposed approach can achieve a high accuracy of human activity recognition, even considering people with di different heights.