方位依赖幅相误差影响下的非圆信号多站直接定位算法

Multi-station DPD Algorithm for Non-circular Signals in the Presence of Angularly Dependent Gain/phase Errors

基于阵列观测站的目标定位易受阵列误差的影响。针对方位依赖幅相误差影响下的定位问题,提出了一种利用辅助阵元(Instrumental Sensor,IS)自校正的非圆信号多站直接定位(Direct Position Determination,DPD)算法。该算法利用辅助阵元校正方位依赖的阵列幅相误差,结合信号的非圆特性扩展数据模型,基于扩展子空间数据融合准则直接实现多目标位置参数与多站阵列误差参数的解耦估计,避免了传统两步定位方法的测向与数据关联步骤。仿真结果表明,该算法较传统的两步定位方法与未考虑阵列误差的直接定位算法均具有更高的定位精度,尤其在低信噪比与目标距离较近时优势更加明显。

Localization using observer arrays is easily influenced by array errors.To deal with the location problem in the presence of angularly dependent gain and phase errors,a multi-station direct position determination(DPD)algorithm for non-circular signals based on instrumental sensors(IS)is proposed to implement auto-calibration in this paper.The algorithm uses the instrumental sensors to calibrate the angularly dependent gain and phase errors,and combines the non-circular property of signals to extend the data model.It directly realizes the decoupling estimation of multi-target position parameters and multi-station array error parameters based on the extended subspace data fusion criterion,which avoids the process of direction finding and data association inherent in the traditional two-step location methods.Simulation results show that this algorithm has higher positioning accuracy than the traditional two-step positioning method and the DPD algorithm without considering array errors,especially in the low signal-to-noise ratio region or when the targets are closely located.