Multi-level radio tomographic imaging based three-dimensional static body posture sensing

This article presents a preliminary research on radio tomographic imaging (RTI) based approach for three-dimensional static body posture sensing. A wireless network organized by a multi-level of radio frequency (RF) sensor array is introduced for posture sensing covering a three-dimensional space. It is assumed the statistical shadowing losses on the passing links between pairs of nodes will be attenuated by the occlusion body. Then an attenuation tomographic image of body posture can be obtained by using the received signal strength (RSS) measurements. Considering the property of spatial piecewise constant of body, total variation (TV) minimization is used to reconstruct the three-dimensional gray image of posture. Experimental studies show the proposed method is able to reconstruct three-dimensional body with several kinds of posture, which will bring significant benefits for future behavior analysis as well as many other applications.