论文部分内容阅读
一、引言本文利用多重假设测试(MHT)理论来处理目标的分类,这种理论是统计判定理论的一个分支。把要分类的这些目标分成许多类别。分类机自地把把进入的被接收信号识别为属于其中一个类别。通过由一些有代表性的参数所取的数值来辨别各个类别。所使用的参数数目越大,则这些目标类别之间的区分就越可靠。在雷达情况下可利用的一组参数是雷达横截而(RCS)、多卜勒频谱的形状和极化状态。测定这些参数需要有相干极化雷达。利用目标的角频谱对目标分类可能也是值得做的,在这种情况下,传感器应该装有相控阵天线,以便在角量度域内获得信号取样。这种方法的推广导致了稀疏阵列的概念,并促使利用传感器组网来
I. INTRODUCTION This article uses the Multiple Hypothesis Testing (MHT) theory to deal with the classification of objects. This theory is a branch of statistical theory. These goals to be classified fall into many categories. The classifier recognizes the incoming signal itself as belonging to one of the categories. Each category is distinguished by the value taken by some representative parameter. The greater the number of parameters used, the more reliable the distinction between these target categories. A set of parameters that are available in the radar case is the radar cross section (RCS), the shape and polarization of the Doppler spectrum. Determination of these parameters requires coherent polarization radar. It may also be worthwhile to use the angular spectrum of the target to classify the target, in which case the sensor should be equipped with a phased array antenna in order to obtain signal samples in the angular measurement domain. The promotion of this method has led to the concept of sparse arrays and prompted the use of sensor networks