基于运动捕获文件的人体微多普勒信号仿真

针对传统人体微多普勒信号仿真算法适应模型简单、与人体相似度不高的问题,提出了一种基于运动捕获文件的仿真方法。该方法从运动捕获文件中获得各关节点的位置随时间的变化,根据关节点的相对位置建立由12个椭球体组成的人体三维模型,并根据此模型仿真RCS回波。此方法所建模型与实际人体外形相似度高,可对任意运动模式的人体雷达回波进行仿真。仿真结果验证了所提算法的正确性。

Utilizing motion captured file to simulate micro-Doppler signal of human beings

Tradition methods that simulate the micro-Doppler signal echoes of human motion for radars can only suit for simple motion model and the established human model has little similar with real human figure. Aim to this, this paper proposed a method based on the motion captured file. At first, it calculated the position transformation of articulations from the motion files. Subsequently, established a three dimension human model, which was composed by twelve ellipsoids, base on the positions of articulations, and simulated the RCS of human body based on the established model. The proposed method can calculate the echoes for arbitrary model of human motions; also, the established model is similar with human figure. Simulation results show the efficiency of the method.