海杂波背景下基于RBF神经网络的目标检测

对海杂波背景下雷达目标检测的最新研究表明,海杂波具有混沌的许多典型特征.本文利用海杂波具有混沌行为这一先验信息,构造了一个神经网络预测器来重构海杂波的内在动力学,并引入一种基于混沌的检测方法对Swerling I型目标和雷达采集的实际海杂波数据进行检测分析,同时讨论了嵌入延迟τ对检测性能的影响.实验结果表明,这种检测方法能有效地实现海杂波背景下的目标检测,并且其检测性能随τ的增大呈下降的趋势.

Target Detection in Sea Clutter Based on RBF Neural Network

Recent work of radar target detection in the background of sea clutter shows that sea clutter has many chaotic properties. In this paper, a neural network predictor is constructed to reconstruct the underlying dynamics of sea clutter which is known to be chaotic. Detection analysis using Swerling I target and real sea clutter data sets collected by radar is carried out by use of a chaos-based detection method. At the same time, the influence of the embedding delay r on the performance of this detection method is studied. Experimental results show that the method can detect the target embedded in the sea clutter background effectively. In addition, increasing the embedding delay r has a negative impact on the performance of the method.