论文部分内容阅读
针对盖氏圆信源数目估计方法在低快拍数、低信噪比情况下性能变差的问题,设计了一套能在有色噪声中提高估计准确率的方法。首先,对接收信号样本集多次抽样,利用多个子样本集协方差矩阵的行列变换构建不同酉矩阵,然后对多个酉变换结果取平均,最后通过多次重采样循环迭代得到最优估计值。仿真结果表明,相比传统GDE的统计方法,文中提出的方法能得到更稳定、准确的估计结果,同时计算所需的接收信号快拍数目更少、信噪比更低。
Aiming at the problem that the method of estimating the number of Galerkin sources has poor performance at low snapshots and low signal to noise ratio, a method of improving the estimation accuracy in colored noise is proposed. Firstly, multiple samplings of the received signal samples are used to construct different unitary matrices by using matrix transformation of multiple sub-sample covariance matrices, and then averaging multiple unitary transform results. Finally, the optimal estimation value is obtained through multiple re-sampling loop iterations . The simulation results show that the proposed method can get a more stable and accurate estimation result than the traditional GDE method, and the number of receive signals needed for the calculation is less and the SNR is lower.