Research Of Parameters Estimation And Filtering With Polarization Sensitive Arrays

Polarization is important information of electromagnetic source besides time and space domain information, which is of great significance in target detection and recognization, signal improvement, polarization filtering. But polarization is far from being thoroughly studied and exploited. Parameters including polarization estimation based on polarization sensitive vector sensors and polarization filtering is studied in this doctoral dissertation. The primary contributions included in this dissertation are listed bellow:1. In chapter 3, algorithms of parameters estimation of completely polarized (CP) transverse electromagnetic waves is provided.(1) A novel algorith for estimation of the angles of arrival and polarization using dipole triads or loop triads is proposed in this paper. The method can resolve the indetermination of quadrant. The error of reference was analyzed.(2) A new method for estimation of the direction of arrival (DOA) based on arbitrary L-shaped electromagnetic vector sensor (VS's) array is proposed. A VS's consists of six co-located but diversely polarized antennas separately measuring all six electromagnetic-field components of an incident wavefield. The new makes use of the array aperture information and improves the estimation accuracy by an iterative algorithm.(3) A novel method for joint frequency, 2-D direction-of-arrrival and polarization estimation of transverse electromagnetic waves based on arbitrary L-shaped VS's array is proposed. Sub-Nyquist temporal sampling is introduced to reduce the complexity of the system design. Furthermore, sub-Nyquist spatial sampling, which can further reduce the mutual coupling among VS's, is combined to improve the estimation performance of spatial parameters. The CRLB is also derived in the end of this paper to verify the efficiency of this method.(4) A method of joint estimation of the frequency, angle of arrival and polarization with crossed dipoles was investigated in section 3. The proposed method overcomes the disadvantage of most existing methods in which only the eigenvector of the forth-order cumulant matrix is used, thereby greatly reduce the computation burden.2. In chapter 4, methods for estimation parameters of partially polarized transverse electromagnetic waves are provided.(1) A new method was proposed to estimate the DOA and degree of polarization (DOP) of partially polarized (PP) electromagnetic waves. Considering the property of electromagnetic waves, a new means to obtain correlation matrix was given and the new method reduce the request of DOA of signal.(2) A new method was proposed to estimate the parameters PP electromagnetic waves. This method makes full use of the eigenvalues and eigenvectors of matrix. Parameters can automatic pair and greatly reduce computation and highly improve precision.3. In chapter 5, methods of caliberation and remedy a VS is proposed.An ESPRIT-based novel method for calibration and remedy a VS with gain and phase uncertainties is presented in this paper. Two known electromagnetic sources with same DOA are used. The computational formulas for gain and phase uncertainties and the CRLB of every parameter are derived.4. In chapter 6, methods of polarization filtering is given.(1) A novel algorithm for joint filtering in spatial, spectral and polarizational domains with a VS is proposed. It makes full use of the information in three domains and signals can be resolved from the interference background provided there is separation between signals and interferences in any domains.(2) To improve the performance of polarization filtering, the robust beamform method was used in polarization filtering.