A K-means Clustering and Triangulation-Based Scheme for Accurate Detection of Multiple Adjacent Through-the-Wall Human Targets.

Through-the-wall (TTW) human targets' detection is extensively desired in civil and military applications. Most studies in this field have only focused on multiple human targets with large spacing, while the situation of short spacing has yet to be well-treated. To solve this problem, an accurate and robust postprocessing scheme is developed in this article for detecting multiple adjacent TTW human targets. First, we identify human targets in the range slow-time planes with a high detection rate. Second, according to the comparing result between different operating planes, two localization solutions based on K-means clustering and triangulation are proposed to extract the precise spatial positions of human targets. Third, using the designed radar system, the effectiveness of the proposed scheme is verified by two typical simulations and three field experiments. The results indicate that the proposed scheme can accurately detect multiple TTW human targets under conditions including large, short spacing, and different orientations with a localization accuracy of 20 cm.