Blind Separation of Radar Signal Based on the Estimation of AR Model's Order
Jian-Xiong Wang, Limin Zhang, Zhaogen Zhong · 2011
Most of the blind algorithms are based on one of the following properties: nongaussianity of the sources, their different autocorrelations, or their nonstationarity. Each of the methods is able to separate sources if the respective assumptions are met. But in reality, radar signals fulfill all of the three conditions at the same time. So we adopt the autoregressive model. And through algorithm based on Maximum Likelihood we can separate independent sources if any of these conditions is met. From the result of simulation, the correlation coefficient between estimate signals and sources were larger than 0.99, and the global error was confined to permission.