Contactless Sensing Using Intelligent Walls

Human activity monitoring is a fascinating field of study that can help the disabled and/or elderly patients to live independently. Different ways have been proposed to identify human activities, including using sensors, cameras, wearables, and non-contact microwave sensing. Microwave sensing has recently attracted a great deal of attention because of its ability to resolve the privacy problems associated with cameras and the discomfort produced by wearables. Existing microwave sensing approaches, however, have the fundamental problem of requiring regulated and perfect conditions for high-accuracy activity detections, which limits their widespread application in non-line-of-sight (Non-LOS) contexts. Intelligent wireless walls (IWW) are proposed to enable high-precision activity monitoring in complicated areas where standard microwave sensing is ineffective. The IWW is a reconfigurable intelligent surface (RIS) capable of beam steering and beamforming and incorporated with machine learning algorithms, can accurately and automatically recognise human behaviours. Two complicated environments were considered for the experiment: a corridor junction scenario in which the transmitter and receiver are located in distinct corridor sections and a multi-floor situation in which the transmitter and receiver are located on various building levels. Three separate bodily movements are evaluated in each of the environments: sitting, standing, and walking. Two individuals, one male and one female, performed these tasks in both scenarios. It is shown that IWW provides a maximum detection increase of 28% in a multi-floor situation and 25% in a corridor junction scenario when compared to conventional microwave sensing without RIS.