一种空间目标RCS序列的特征提取方法

New Feature Extraction Method for RCS Sequence of Space Target

准确、快速判断空间目标的姿态运动模式,对于空间目标监测具有重要意义.针对空间目标雷达散射截面(Radar Cross Section,RCS)序列,提出一种基于变分模态分解(Variational Mode Decomposition,VMD)与盒维数的特征提取方法.在频域对空间目标的雷达散射截面序列进行变分模态分解,得到若干本征模态分量,然后计算每个模态分量的盒维数以构造特征向量.采用前向反馈神经网络分类器验证识别效果,讨论了不同轨道高度对识别准确率的影响.结果表明上述方法能有效识别三轴稳定类与旋转类的空间目标,相比传统的统计参数特征识别准确率提高约10%,并且识别率随轨道高度增加呈上升趋势.

Accurately and quickly identifying the attitude motion pattern of space targets is of great significance for space target surveillance. Aiming at the feature extraction for radar cross section (RCS) sequence,a new method based on variational mode decomposition (VMD) and box dimension is proposed. In the frequency domain,the RCS sequence of space target was processed with VMD to obtain intrinsic modes. Then the box dimension of every mode was calculated to construct a feature vector. The recognition effect was verified with back propagation neural network classifier,and the impact of different orbit heights on the recognition accuracy was discussed. The experimental results indicate that the proposed method can effectively identify the three-axis stabilized and rotating space targets,and the accuracy is about 1 0 % higher than the feature extraction method of statistical parameters. As the orbit height increases,the recognition accuracy shows an upward trend.