基于特征融合的雷达信号脉内调制类型识别

针对雷达信号脉内调制识别算法准确率低的问题，提出基于特征融合的雷达脉内调制类型识别方法，该方法首先提取雷达信号时频图像的形状特征和纹理特征，利用改进的主成分分析法（IPCA）对特征进行融合，然后将融合特征输入支持向量机（SVM），实现信号的分类识别。仿真实验中对8种常见的不同调制类型的雷达信号进行识别，该算法在信噪比为5dB时识别准确率接近100％，验证了该方法的有效性。

Intra-pulse modulation recognition of radar signal based on features fusion

To solve the problem of low recognition rate of intra‐pulse modulation recognition of radar signal ,the recognition method based on features fusion is proposed .The method obtains the shape and texture features of time‐frequendy image of radar signal ,integrates the features by the improved principal component analysis (IPCA ) ,and then inputs the fusion features to the support vector machine (SVM ) to identify the signals .The results of simulation experiments of eight kinds of common radar signals show that recognition accuracy of the method proposed can almost achieved 100% in SNR 5dB ,so this method is very effective .