关键词: 高分辨距离像, 极化, 雷达目标识别, 核函数

Abstract:

Aiming at the great quantity of multi-polarization high resolution range profile (HRRP), the complexity of the data distribution and the recognition algorithm, the methods based on kernel methods are proposed. Firstly two kernel functions based on the multi-polarization HRRP are defined,  and then two kernel functions are used to the kernel principal component analysis (KPCA) respectively. Finally, the nearest neighbor (1NN) classifier and the support vector machine (SVM) classifier are used for classifying targets. The multi-polarized radar HRRP can be recognized as a whole in the proposed methods, so the complexity of the recognition algorithm is reduced. Experimental results based on simulated multi-polarization HRRP data show that the methods based on the proposed kernel functions can raise the correct recognition rate greatly compared with the methods of single-polarized HRRP recognition. Also, the feature dimension can be decreased and the recognition performance can be improved to some extent compared with other methods of the multi-polarization HRRP.

Key words: high-resolution range profile, polarization, radar target recognition, kernel function