辐射源识别技术是当前电子侦察的重要研究方向。文中利用雷达模拟器的脉冲包络指纹特征,构建特征数据库,提出一种新的指纹特征向量匹配识别方法。该方法根据自适应评分机制,衡量不同特征向量之间的相似性,实现对辐射源脉冲的分类和识别处理。利用同型号雷达模拟器的外场实录数据,验证了该识别方法的有效性,分类性能优于传统的M-距离法。文中还定量分析了辐射源分类正确率与信噪比的关系,结果表明,分类正确率与SNR成正相关,当信噪比高于20 dB时,分类正确率超过95%。

Based on fingerprint features, special emitter identification (SEI) plays an important role in electronic investigation. In this paper, a database of emitter fingerprint feature is constructed, by extracting pulse envelope fingerprint features of radar simulator. And then a new vector matching method is proposed. According to the adaptive scoring mechanism, the similarity between two vectors can be measured, and pulses can be identified and verified. The effectiveness of the proposed matching method is verified by using experimental data from the same model radar simulators. The classification performance is better than the traditional M-distance method. In addition, the relationship between signal-to-noise ratio (SNR) and the classification accuracy is quantitatively analyzed. The results show that the classification accuracy is positive correlation to the SNR. When the SNR is higher than 20 dB, the accuracy of SEI is over 95 percent.

TN955