| 摘要: |
| 针对射频噪声 距离欺骗加性复合干扰、噪声调幅 距离欺骗加性复合干扰、噪声调频 距离欺骗加性复合干扰、灵巧噪声干扰信号、距离欺骗干扰，建立了目标与干扰模型，仿真了其干扰效果。采用基于时域、频域及其他域的多维特征提取方法提取信号特征，然后采用决策树分类器进行分类，最后使用主成分分析法对提取的19维特征因子进行有效降维到9维，仿真实验结果表明：特征因子降维前，在JNR=5dB处，各类干扰信号有85%以上的正确识别率；降维后的识别效果与降维前无明显差异，实现了对冗余数据的有效去除。 |
| 关键词: 雷达加性复合干扰 灵巧噪声干扰 特征提取 主成分分析法 |
| DOI：DOI:10.3969/j.issn.1672-2337.2019.06.002 |
| 分类号:TN974 |
| 基金项目: |
|
| Research on Noise Modulated Active Jamming Signal Recognition Technology |
|
LUO Binshen，LIU Limin，LIU Jingqi，DONG Jian
|
|
Shijiazhuang Campus of Army Engineering University，Shijiazhuang 050003，China
|
| Abstract: |
| This paper studies the recognition of five kinds of active jamming signals，including the compound jamming of the range deception jamming signal and RF(radio frequency) noise，the compound jamming of the range deception jamming signal and AM(amplitude modulation) noise，the compound jamming of the range deception jamming signal and FM(frequency modulation) noise，the smart noise jamming signal，and the range deception jamming signal. To deal with the jammings，a recognition method based on multiple features is proposed in this paper.The feature signals are extracted by the multi dimensional feature extraction method based on time domain，frequency domain，and other domains，and then decision tree classifier is used to recognize. Finally，the PCA(principal component analysis) is used to reduce the 19 dimension data to 9 dimensions.The simulation results show that when the JNR is 5 dB，the recognition rate of each type of signal is greater than 85%，the recognition effect is retained，and the effective removal of redundant and repetitive signals is achieved. |
| Key words: radar composite jamming signals smart noise jamming feature extraction principal component analysis (PCA) |
