Source number estimation under colored noise based on resampling strategy
Yang Jin-shen · Information technology newsletter · 2015
In order to solve the problem that the performance degradation of gerschgorin disk source number estimation method in the condition of small snapshots and low SNR,this paper designs a set of strategy to improve the detection accuracy in color noise. Resampling theory and multiple unitary transformations averaging are combined to define the optimization estimation result by employing subsets of the whole data set iteratively. Different unitary matrices are constructed from covariance matrix of sample subset. The simulation results indicate that,the strategy greatly improves the performance of the existing gerschgorin disk detection method by making full use of the limited available data with lower SNR.