Through-Wall Localization Based on Exponential Moving Average and Distance Decay Elliptic Model

As a new and important technology in wireless sensor networks, wireless tomography has been widely concerned by researchers at home and abroad. However, the target positioning accuracy in radio tomographic imaging (RTI) still has a lot of room for improvement. In indoor environments, signals need to be propagated through the wall and reflected through the wall, which will produce multipath effect and affect the positioning accuracy of the target. In this paper, a data preprocessing method and an improved range-attenuated elliptical weight model are proposed to further improve the target positioning accuracy. The data preprocessing method selects exponential moving average algorithm to process the received signal strength (RSS), which can effectively reduce the multipath effect. The new elliptic model of distance attenuation considers the distance between pixels and sensor nodes. The effects of line-of-sight (LOS) and non-line-of-sight (NLOS) paths are also considered. Experimental results show that the root mean square error of position can be controlled at 0.31m.