论文部分内容阅读
在基于自适应噪声完备集合经验模态分解方法(CEEMDAN)提取信号特征的基础上,结合加法核支持向量机(ASVM)实现对低截获概率雷达信号的识别.首先根据CEEMDAN分解得到本征模态函数(IMF);然后依据信号的IMF分量特点,提取频率方差、幅度样本能量熵、灰色相似度三种特征;最后利用ASVM对低截获概率雷达信号进行识别,提高识别速度.理论分析和仿真实验结果表明,该模态分解法对信号特征的识别有较强的鲁棒性;在信噪比大于5 dB时,算法对四种雷达信号的识别率能达到80％以上;在识别准确率上,与选用的四种分类算法比较大致相同,但有最快的识别时间.,A novel recognition algorithm for low probability of intercept (LPI) radar signal based on additive kel support vector machine (ASVM) of radar signals’complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is proposed in this paper.Firstly,a series of intrinsic mode function (IMF) can be decomposed by CEEMDAN.Secondly,according to the characteristic of IMF,three features,such as frequency variance,sample energy entropy and grey similarity,were extracted.Finally,the ASVM is used to classify and recognize LPI signals.The theoretical analysis and the simulation results show that this empirical mode decomposition method has strong robustness.The recognition accuracy of the proposed algorithm can reach 80％ or ever higher for the four kinds of LPI radar signal and comparing to four classification algorithms,the algorithm has the best recognition time.