Source localization based on the array aperture loss for channel phase calibration
Source localization based on the array aperture loss for channel phase calibration
复制标题
DOI：
10.1002/dac.5279
复制
发表时间：
2022-06
影响因子：
2.1
通讯作者：
Weiwei Hu;Guangwei Qing
中科院分区：
文献类型：
--
作者：
Weiwei Hu;Guangwei Qing
提出了一种在信道相位误差存在的情况下，利用均匀圆阵（UCA）实现信源定位的新方法。受空间域中的降秩（RARE）理论的启发，该理论利用仪器传感器创建必要条件（降秩向量），基于贝塞尔函数的阶数和自变量之间的关系，考虑减小UCA的孔径以构造仅与UCA模式域中的源位置相关的降秩向量。这表明在UCA模式域中可以解耦源位置和信道相位误差，然后利用基于RARE理论的降秩向量来构造仅与信道相位误差相关的优化函数。然后，校准的信道相位误差，源的位置可以估计与Vandemonde结构的虚拟导向矢量通过UCA-ESPRIT。所提出的方法不需要预先校准的传感器或校准源。此外，它避免了多维迭代。通过源分离、信噪比（SNR）和快拍实验验证了该方法可以近似地实现标定源法和仪器传感器法;此外，还验证了所提出的方法可以科普比基于Schur-乘积的方法更多的源，并且由于其对阵列孔径具有更大的容差，因此在性能上优于基于Schur-乘积的方法上级，而这两者都是在线估计器
A novel method is proposed to locate sources with uniform circular array (UCA) in the presence of channel phase errors. Inspired by the rank reduction (RARE) theory in the spatial domain, which creates the necessary condition (reduction rank vector) with the instrumental sensors, a reduction in the aperture of UCA is considered for the construction of the reduction rank vector that is related only to the source location in the UCA‐mode domain based on the relationship between the order and the independent variable of the Bessel function. This indicates that the source location and the channel phase errors can be decoupled in the UCA‐mode domain, and then the optimization function that is only related to the channel phase errors is formulated with the reduction rank vector based on the RARE theory. Then with the channel phase errors calibrated, the source location can be estimated with the Vandemonde structure of the virtual steering vector via UCA‐ESPRIT. The proposed method requires no pre‐calibrated sensors or calibration sources. Also, it avoids the multi‐dimensional iteration. Results verify that the proposed method can perform approximately to calibration source method and instrumental sensor method via source separation, signal‐noise ratio (SNR), and snapshots experiments; furthermore, it is also verified that the proposed method can cope with more sources than the Schur‐product based method and it is superior to the Schur‐product based method in performance as it has a larger tolerance for the array aperture, while both of them are on‐line estimators.