基于3D-Hibert能量谱和多尺度分形特征的通信辐射源个体识别

发布时间：2019-01-11 19:08

【摘要】：针对通信辐射源的个体识别问题,提出一种基于希尔伯特—黄变换(HHT,Hilbert-Huang transform)和多尺度分形特征的新方法。首先,通过HHT得到时频能量谱,将其视为三维空间中的复杂曲面,即3D-Hilbert能量谱;然后,利用分形理论通过多尺度分块提取差分盒维数和多重分形维数二维特征组成特征向量;最后,采用支持向量机分类器结合二维特征向量实现通信辐射源的个体分类。分别利用仿真信号和调制方式相同的实际通信信号,验证并对比了所提方法与另外2种方法在2类及3类目标情况下的识别性能。实验结果表明,所提方法的识别率远高于其他2种方法,能够克服低信噪比和少训练样本数量对识别性能的负面影响,证明了所提特征的稳定性、充分性及可分性。
[Abstract]:A new method based on Hilbert-Huang transform (HHT,Hilbert-Huang transform) and multiscale fractal features is proposed for individual identification of communication emitter. Firstly, the time-frequency energy spectrum is obtained by HHT, which is regarded as a complex surface in three-dimensional space, that is, 3D-Hilbert energy spectrum. Then, fractal theory is used to extract the two-dimensional feature vectors of difference box dimension and multifractal dimension by multi-scale block. Finally, support vector machine classifier combined with two-dimensional feature vector is used to realize individual classification of communication emitter. The performance of the proposed method is verified and compared with that of the other two methods in the case of two or three kinds of targets by using the simulation signal and the real communication signal with the same modulation mode respectively. The experimental results show that the recognition rate of the proposed method is much higher than that of the other two methods. It can overcome the negative effects of low SNR and the number of training samples on the recognition performance, and prove the stability, adequacy and separability of the proposed features.
【作者单位】：解放军信息工程大学信息系统工程学院;
【基金】：国家自然科学基金资助项目(No.61572518)~~
【分类号】：TP391.41

【参考文献】