Using a Kinect WSN for home monitoring: Principle, network and application evaluation

Tracking people and gesture recognition become an application increasingly exploited for control and supervision especially with the invasion of networks and communication systems. In this paper we propose a wireless sensor network using Kinect video sensor to control and supervise the state of ill or old person into a smart home. We discuss the use of a WiFi network with mesh topology and the application of a routing protocol like OLSR or 802.11s standard. For each node, we investigate the response time and the link quality. We try to transport Kinect data in our network using USB over IP encapsulation. Results show that it is difficult to disseminate data from the sensor in the mesh WiFi network because of the high flow rate required which is about 180Mbits/s and open prospects for using another wireless standard.