论文部分内容阅读
针对宽开通信侦察系统中的多信号分离问题,提出了一种总体平均经验模式分解(EEMD)方法。首先对混合信号进行去噪,进行功率谱估计,然后利用EEMD方法对混合信号进行分解得到一簇本征模态函数IMF(Intrinsic Mode Function),对代表单个信号的IMF进行Hlibert变换,得到与各阶IMF相对应的瞬时频率,确定信号的个数,估计出其载波频率,设计出相应的带通滤波器,分离出单个信号。最后设计决策树对分离出的单个信号进行调制识别,确定信号的种类。文中以三个信号组合ASK、BPSK、2FSK为例,通过仿真验证了方法的有效性。
Aiming at the problem of multi-signal separation in wide open communication reconnaissance system, an overall average empirical mode decomposition (EEMD) method is proposed. Firstly, the mixed signal is denoised and the power spectrum is estimated. Then, the EEMD method is used to decompose the mixed signal to get a cluster of Intrinsic Mode Function (IMF), Hlibert transformation is performed on the IMF representing a single signal, Order IMF corresponding to the instantaneous frequency, determine the number of signals to estimate the carrier frequency, design a corresponding band-pass filter, separate a single signal. The final design decision tree to separate the single signal modulation recognition, to determine the type of signal. In this paper, three signal combinations ASK, BPSK and 2FSK are taken as an example to verify the effectiveness of the method.