基于滤波器组和高阶累积量技术的LPI信号特征检测的新方法  

Research on LPI signals feature detection based on parallel filter bank and higher order cumulants

作　　者：戴幻尧[1] 蒋鸿宇[2] 

机构地区：[1]国防科技大学电子科学与工程学院,湖南长沙410073 [2]中国工程物理研究院电子所,四川绵阳621900

出　　处：《系统工程与电子技术》2009年第6期1336-1340,共5页Systems Engineering and Electronics

摘　　要：提出了一种新的将滤波器组和高阶累积量相结合的特征检测方法。运用该方法可以对未知LPI多分量信号进行频域的分离并消除伴随的加性高斯和对称非高斯噪声的干扰,从而较好地估计出FMCW、多相编码等LPI信号的特征参数。同时对这种改进的方法进行了系统结构的设计和详细的数学描述。仿真结果和性能分析表明了该方法的有效性和适应性,即使在信噪比很低的情况下也能检测得到比较理想的LPI信号特征。A novel method for detecting the features of LPI signals is presented in which a filter bank is combined with higher order cumulants. The significance of this technique is that it separates the LPI waveform in small frequency bands, thus providing a detailed time-frequency description of the unknown signal, and it sig- nificantly reduces the interference by add -Gaussian noises and symmetric noises so as to estimate the LPI signals feature parameter, such as FMCW, P4, accurately. The paper also gives the design of system structure and de- tailed mathematic description. Finally simulation results and performance analysis illustrate that the proposed method is effective and practical, even if working at the signal environment of low SNR, the detection of LPI signal feature and identification based on the modulation parameters detected is good.

关 键 词：滤波器组 高阶累积量 低截获概率信号 特征检测 参数估计 

分 类 号：TN911.7[电子电信—通信与信息系统]