Research On Polarization Sensitive Array Polarization-Space Spectrum Estimation Algorithm

As passive radar systems based on spatial spectrum estimation techniques can simultaneously estimate direction of arrival(DOA)of multiple signals,it has been widely applied in the fields of electronic reconnaissance and countermeasure with extensive attentions from scholars,engineers and technicians.Recently,the multi-mode composite guidance system has become one of the most important development directions of the precision-guided weapons.In order to be compatible with other conventional sensors used in other guidance modes,a polarization sensitive array has been introduced into the passive direction finding system.The passive direction finding system which is traditionally based the scalar arrays hereby evolves to be based on the polarization sensitive arrays.The polarization sensitive array can effectively improve the estimation performance of direction finding system by estimating the spatial information and the polarization information of the incident electromagnetic wave simultaneously.The polarization sensitive uniform circular array composed of single dipoles and orthogonal dipoles are the typical polarization sensitive arrays applied in the passive direction finding system.Therefore,this dissertation focused on the pratical system problems and study the polarization-spatial spectrum estimation techniques based on two typical polarization sensitive arrays.The main contents of this dissertation include the source number estimation based on these two typical polarization sensitive arrays and a spatial spectrum estimation algorithm with low complexity under the background of pratical engineering applications.The main work of this dissertation is listed as follows.1.A source number estimation algorithm based on the feature eliminated process using polarization sensitive arrays with a single dipole: the spatial spectrum estimation algorithm is based on the premise of estimating the source number accurately.But in the pratical direction finding systems,due to the disturbance of noise received by the array,the probability of estimating the source number decreases.To solve this problem,a source number estimation algorithm based on the polarization sensitive array with a single dipole was proposed in the colored noise environment.Based on the feature eliminated process,this new method of source number estimation enlarged the difference between the covariance matrices respectively corresponding to the signal subspace and noise subspace.With the calculation of the projections belonging to these two matrices on the whole space,a criterion function was contructed to realize the estimation of the source number.The validity of the new method was demonstrated by the computer simulation.2 A source number estimation algorithm based on feature eliminated process usingorthogonal dipoles polarization sensitive arrays: In order to improve the performance of the source number estimation,a new algorithm based on the polarization sensitive array with orthogonal dipoles was proposed.Combining the source number estimation algorithm based on feature eliminated process with the dimensional characteristisc of the data received by the polarization sensitive array composed of orthogonal dipoles,this new method constructed the projection matrix using part of the noise subspace vector and then solved the projection of the new covariance matrix under the feature eliminated process.Then a new criterion function was constructed to estimate the number of sources.Theoretical analysis proved that this new algorithm improved the performance of the source number estimation algorithm based on feature eliminated process by reducing the computation complexity,taking full advantage of the properties of polarization sensitive array and equivalently increasing the quantity of information.Finally,simulation results verified the effectiveness of the proposed algorithm.3 Half-interval search MUSIC algorithm based on polarization sensitive circular array composed on single dipoles: Signals are received in vectors by the polarization sensitive array composed of single dipoles and then these received vectors are merged according to the antenna pointing angle of each array element.In this case,additional phase generated by the polarization sensitive array is so different as to change the phase difference between the array elements,correspondingly undermine the performance of the spatial spectrum estimation technique based on the scalar array or even cause failure.To solve this problem,a mathematical model of the polarization sensitive array in practical engineering applications was constructed to obtain the passive direction finding system model based on the polarization sensitive array composed of single dipoles.The effect on the phase difference between array elements by polarization reception and location of array elements was analyzed by the theoretical deduction and computer simulation.Based on the characteristics of the phase difference,a new half-interval search MUSIC(HIS-MUSIC)algorithm was proposed by constructing a new spectral function.Following the proposed algorithm,half of the four-dimensional space was searched to determine the real signals from the searching results.Then the joint estimation of the direction of arrival(DOA)and polarization of the incident signals by searching half of the four-dimensional space was obtained.Compared with traditional four-dimensional searching algorithms,the proposed algorithm can reduce half of the computational quantity.The efficiency of the proposed algorithm was demonstrated by the computer simulation experiments.4.ESPRIT algorithm based on four-order cumulant using polarization sensitive array composed of orthogonal dipoles: Due to the huge computational complexity of the jointestimation based on DOA by the polarization sensitive uniform circular array composed of orthogonal dipoles and polarized parameters,an ESPRIT algorithm based on four-order cumulant was proposed.In this algorithm,a virtual array extension property of four-order cumulant was yielded to extend the polarization sensitive uniform circular array composed of the orthogonal dipoles,then to construct pairs of sub-arrays with rotation invariance and thereafter to jointly estimate polarization domain-spatial domain information of the incident signals based on ESPRIT algorithm.The proposed algorithm is not dependent on calculating and searching the spectral function in the four dimensional space but to dramatcally reduced the computation quantity of jointly estimating DOA and polarization by using polarization sensitive uniform circular array composed of orthogonal dipole.The computer simulation experiments verified the effectiveness of the proposed algorithm.