Multiple Human Localization Method Using Compressed Sensing

Abstract

This paper investigates the multiple human localization using Multiple-Input Multiple-Output (MIMO) sensor. We propose a living body position estimation method based on a compressed sensing using a newly developed two-stage virtual array technique for enhancing the sensitivity to the vital sign. The proposed method estimates the positions of multiple objects without the knowledge of the number targets by using the correlation matrix obtained from the frequency response of the channel matrix. The experimental results show that the proposed method can estimate the location of a single target with 0.15 m accuracy on average. Moreover, the average accuracies with two and three subjects are 0.25 m and 0.29 m, respectively.