论文部分内容阅读
针对复杂体制雷达辐射源识别,提出一种基于时频分布Rényi熵的雷达信号特征提取和识别方法。该方法首先对雷达辐射源信号进行时频变换,然后提取信号时频分布的3阶、7阶和11阶Rényi熵作为特征向量,得到具有维数低、类间差异较大的识别特征。最后采用支持向量机分类器实现信号的分类识别。文中对8种常见雷达信号进行了仿真实验,结果表明在较大的信噪比范围内,该方法能获得较为满意的正确识别率,当信噪比为-3dB时,采用时频分布Rényi熵特征的平均识别率仍能达到90.75%,验证了提出方法的有效性。
Aiming at the complex system radar emitter recognition, a method of radar signal feature extraction and recognition based on Rényi entropy with time-frequency distribution is proposed. Firstly, the Rényi entropy of the 3rd, 7th and 11th order of the time-frequency distribution of the signal is extracted as the eigenvector by time-frequency transformation of the radar emitter signal, and the recognition features with low dimension and large differences between classes are obtained. Finally, SVM classifier to achieve the classification of signal recognition. In this paper, eight kinds of common radar signals are simulated, and the results show that this method can obtain a satisfactory correct recognition rate in a large range of signal to noise ratio. When the SNR is -3dB, the time-frequency distribution Rényi entropy The average recognition rate of features still can reach 90.75%, which verifies the effectiveness of the proposed method.