Direction Finding Technology For Near-Field Sources Based On The Polarization Sensitive Array

Polarization sensitive array is a new type of array which can obtain polarization information of electromagnetic wave signal.Compared with the traditional scalar array,polarization sensitive array has higher resolution and better anti-interference.On the one hand,For the complete polarization sensitive array element,six electromagnetic components of the electromagnetic wave signal can be obtained,and the data dimension is six times that of the scalar array,which greatly increased the complexity of signal modeling and the computational complexity of signal processing.On the other hand,considering that in many practical applications,the source is often located in the nearfield of the array,resulting in a performance degradation or even failure of the directionfinding algorithm based on the far-field hypothesis.Therefore,the near-field source parameters estimation has always been a difficult point in the field of array signal processing,especially based on special polarization sensitive array.Focusing on this difficulty,this paper lays emphasis on how to improve the parameter estimation accuracy of polarization sensitive array by using the non-circular characteristics of signals.At the same time,two high resolution direction-finding algorithms based on quaternions were proposed.The main innovations of this paper as follow:1.Near-field polarization source localization algorithm based on non-circular characteristics was proposed: NCRARE(Non-Circular Rank Reduction)algorithm.The algorithm uses the non-circular signal ellipse covariance matrix to be non-zero to extend the received signal of the array and achieve the purpose of expanding the array virtual aperture and improving direction-finding accuracy of the algorithm.Meanwhile,the traditional polarization sensitive array often requires two-dimensional search to estimate the polarization information of the sources,which means large calculation amount.However,after using the non-circular characteristic of the signal and the special structure of polarization sensitive array,the polarization parameters and noncircular phase of the source were directly estimated by solving the form of the quaternary equations.This method greatly reduced the complexity of the algorithm and improved the accuracy of the parameter estimation.2.Two polarization sensitive array direction finding algorithms based on quaternion were proposed: QNC(Quaternion Non-Circular)algorithm and AQRARE(Augmented Quaternion Rank Reduction)algorithm.The QNC algorithm used the noncircular characteristic of the signal,and the AQRARE(Non-Circular Rank Reduction)algorithm used the symmetry characteristic of the quaternion to extend the dimension of the received signal.The dimension of the covariance matrix was equivalent to the algorithm of the long vector,thereby ensuring the accuracy of the parameter estimation.The quaternion matrix eigenvectors have stronger constraint than the corresponding complex domain.Thus,more accurate signal subspace and better robustness can be obtained than the long vector method.In summary,this paper had made important research and algorithm improvement on the modeling and parameter estimation of polarization sensitive array signal processing.The simulation experiment also proved the effectiveness and advantages of the proposed algorithms.