【摘要】 针对相干准平稳信号、相关噪声环境,提出一种基于Khatri-Rao(KR)子空间和前后向空间平滑的波达方向估计方法。该方法对接收信号进行分帧,并假设各帧内信号是平稳的,通过帧内信号的协方差矩阵构造新的Toeplitz矩阵;通过KR子空间方法将各帧Toeplitz矩阵组合成新的信号模型,该模型可抑制噪声的影响;通过前后向空间平滑求解出信号的来向;同时,方法也扩展到了宽带阵列信号模型。实验结果表明,在准平稳相干源与相关噪声的条件下,所提方法具有更高的检测概率和更低均方根误差,宽带阵列信号模型实验也验证了该方法的有效性。更多还原

【Abstract】 A direction of arrival estimation method based on Khatri-Rao( KR) subspace and forward/backward spatial smoothing was proposed for quasi-stationary coherent signals and correlated noise. The observed signals were divided into frames. Assuming that the signal of each frame was stationary,a new Toeplitz matrix was constructed by the corresponding covariance matrix of each frame. A new signal model which could eliminate the noise was built through the KR subspace method,and the directions of source signals were estimated by forward/backward spatial smoothing method. Meanwhile,the proposed method was expanded to suit the wideband array signal model. The experiment result shows that the proposed method has a better probability of detection and less mean root square error under the conditions of quasi-stationary coherent signals and correlated noise. And the effectiveness of the proposed method is verified by the wideband array signal model experiment.更多还原