论文部分内容阅读
为了提高海面雷达对海漂浮小目标检测能力,提出了一种基于目标时频特征的检测算法。首先,验证了从接收的时间序列中提取的相对多普勒峰高、相对多普勒偏移、相对多普勒熵等特征在时间维度上可以有效地区分海杂波和小目标;其次,构造了特征检测器,给定虚警概率下的判决区域由凸包算法确定;最后,根据实测数据对算法进行了检验。结果表明:当虚警概率为0.01时,采用双特征检测器可在512个脉冲下完全区分海杂波和实测小目标,双特征检测算法优于传统动目标检测算法。
In order to improve the ability of sea surface radar to detect small floating targets, a detection algorithm based on target time-frequency features is proposed. Firstly, it is verified that the characteristics of relative Doppler peak, relative Doppler shift and relative Doppler entropy extracted from the received time series can effectively distinguish sea clutter and small target in the time dimension. Secondly, The feature detector is constructed. The decision area under the given false alarm probability is determined by the convex hull algorithm. Finally, the algorithm is tested based on the measured data. The results show that when the false alarm probability is 0.01, the dual feature detector can completely distinguish the sea clutter and the measured small target under 512 pulses. The double feature detection algorithm is superior to the traditional moving target detection algorithm.