基于改进Dijkstra算法与时频域滤波的雷达目标识别

针对弹道中段雷达目标回波的微多普勒特征提取精准度不高导致目标识别率低的问题, 提出一种基于改进Dijkstra算法与时频域滤波相结合的雷达目标分类识别方法。该方法首先采取改进Dijkstra算法提取多分量回波信号中最强分量的瞬时多普勒特征, 然后利用时频域滤波方法滤除最强分量, 依次提取多分量信号的瞬时多普勒特征, 并将该特征应用于弹道中段雷达目标识别。仿真结果表明, 该方法适用于多种微动形式, 提取回波信号的微多普勒特征的精度更高, 对于弹道中段雷达目标平均识别率较高。

Radar target recognition based on improved Dijkstra algorithm with time-frequency domain filtering

An identification method of radar targets based on improved Dijkstra algorithm with time-frequency domain filtering is proposed to solve the problem that the accuracy of micro-Doppler feature extraction of radar targets in the middle of the trajectory is not high, resulting in low target recognition rates. The improved Dijkstra algorithm is used for extracting the instantaneous Doppler features of the strongest component in the multi-component echo signal. Then the strongest component is filtered through the time-frequency domain filter. So the instantaneous Doppler features of the multi-component signal can be extracted in proper sequence, and features are applied to identify radar targets in the middle of the trajectory. The simulation results demonstrate that this method is suitable for a variety of micro-motion forms, the micro-Doppler features of the echo signal have higher accuracy, and the average recognition rate of radar targets in the middle of the trajectory is increased effectively.