基于自适应重采样次数的协方差角度估计算法

A covariance-based DOA algorithm based on adaptive resampling

针对基于Hermitian分解的协方差角度估计算法在低信噪比(SNR)时性能差的问题,提出了基于自适应重采样次数的协方差角度估计算法。该算法首先用接收信号扫描角度空间内的所有导向矢量,划定正确的角度估计区间,并给出判断角度估计结果是否可靠的标准。然后构造协方差矩阵,得到角度估计结果,接着以角度估计结果是否可靠为判断标准自适应地对原始信号进行重采样。若重采样后对应某一信号的多个角度估计结果都不可靠,则对原信号重采样并估计到达角,直到所有信号均有可靠结果为止。仿真结果表明,所提算法修正异常的角度估计结果,提高协方差角度估计算法在低信噪比时的精确度。

Aiming at the problem that Hermitian-decomposition Covariance-based Direction of Arrival(DOA)algorithm performance is poor in the case of low Signal to Noise Ratio(SNR),a Covariancebased DOA algorithm based on adaptive resampling is proposed.The algorithm first scans all the weighting coefficients in the angular space with the received signal to define the correct angle estimation interval,and gives the criterion to judge whether the angle estimation result is reliable.Then,the covariance matrix is constructed,and the characteristics of the covariance matrix are decomposed to obtain the angle estimation result;the reliability of the angle estimation result is determined.And then resample the original signal adaptively according to the judgment standard with the angle estimation result being reliable.If the multiple estimation results corresponding to a certain signal are unreliable after resampling,the original signal is resampled and the angle of arrival is estimated until all the signals have reliable results.The simulation results show that the proposed algorithm can completely correct the anomaly angle estimation result and greatly improve the accuracy of the covariance angle estimation algorithm at low SNR.