Study On Passive Direction Finding Technology For Simultaneous Multiple Signals

For radar reconnaissance system,the directions of radar,which cannot change rapidly with time and have stable spatial distribution,can provide important guidelines for signal sorting and radar type recognition.Hence,advanced passive direction finding technology is of great importance for radar reconnaissance system.With more and more electronic equipment applied for modern battlefield,electromagnetic environment becomes more and more complicated and changeable.For passive direction finding subsystem,the problem that multiple signals belonging to the same frequency band arrive at the same time cannot be avoidable.In terms of traditional direction finding methods,such as monopulse,interferometer and time difference,they can deal with simultaneous multiple signals to some extent via some special methods.However,they cannot handle simultaneous multiple signals belonging to the same frequency band.Therefore,to improve the interception probability and angle estimation performance of direction finding system,the research on passive direction finding for simultaneous multiple signals and even simultaneous multiple signals belonging to same frequency band is of great importance for modern electronic reconnaissance.Utilizing the direction of arrival(DOA)estimation technique,which has been developed in recent decades and can solve the problem of direction finding in the presence of simultaneous multiple signals and even simultaneous multiple signals belonging to same frequency band,and with the practical requirement of high precision and high resolution passive direction finding subsystem,several two-dimensional DOA estimation methods for L-shaped array are developed in this dissertation.The main work can be summarized as follows:In the first part,three kinds of two-dimensional(2-D)direction finding algorithms for Lshaped array,which can reduce the computational complexity,have been proposed progressively based on the spatial distribution complexity of sources to estimate.In terms of the first algorithm,it firstly utilizes the conjugate symmetry property of the L-shaped array's manifold matrix to extend the array aperture,and then applies the Nystr?m method to reduce the amount of calculation related to signal subspace estimation,finally,the two dimensional angles are estimated via ESPRIT(estimation of signal parameters via rotational invariance techniques)method without peak search.It can greatly reduce the amount of calculations for large arrays,and is particularly suitable for large and very large land-based and shore-based reconnaissance systems.However,the first algorithm may occur blind angle in the procedure of pair matching.To solve this problem,on the basis of inheriting the aperture expansion ability of the first algorithm,the second algorithm estimates the signal subspace with the extended propagator method and removes the angle pair matching and peak search using the principle of ESPRIT method and eigenvalue association technique.It can improves the computational efficiency effectively,be especially suitable for airborne and space-borne reconnaissance systems,and has a wider range of application to the source spatial distribution than that of the first algorithm.For the third algorithm,to solve the problem that the first two algorithms and other low computational complexity algorithms available cannot handle the case that the number of sources is larger than that of the elements in each subarray,it firstly extends the array aperture and array virtual snapshot number simultaneously with the conjugate symmetry properties of the L-shaped array manifold matrix.Then,the angle estimation problem based on 2D-MUSIC algorithm is transformed into the Rayleigh quotient problem by using the characteristics of the augmented array manifold matrix,which reduces the search dimension.Finally,according to the relationship between the resultant eigenvector of the Rayleigh quotient problem and the array steering vector,two dimensional angles are estimated without pairing.The most outstanding advantage of the algorithm is that it can effectively improve the maximum number of identifiable sources and reduce the cost of the system,which can be especially suitable for small missile-borne reconnaissance system.In the second part,to deal with the underdetermined angle estimation problem that the number of sources is larger than that of array elements,two different 2-D direction finding algorithms have been proposed for uniform L-shaped array with uniform linear subarrays and non-uniform L-shaped array with nested subarrays.For the uniform L-shaped array,it firstly extends the aperture and the degree of freedom by using the spatial temporal twodimensional characteristic of the array received data and the conjugate augmented spatial temporal(CAST)technique.Then,the joint diagonalization structure is constructed based on the characteristics of the extended data signal subspace.Finally,the joint diagonalization problem is solved by the Jacobi method with plane rotations.Hence,the underdetermined angle estimation problem is solved.However,in the case of limited aperture,the number of elements required for uniform L-shaped array is too large and the cost is too high.Therefore,the non-uniform L-shaped array with nested subarrays is chosen and its underdetermined angle estimation problem is obtained by using the above-mentioned method for uniform Lshaped array.The advantages of these two methods are that both of them can work in underdetermined condition and have the ability to deal with angle ambiguity problem.The differences between the two are that the latter does not only use the cross-correlation of two subarray data in the CAST procedure,but also uses the autocorrelation of each subarray data,which can extend the array aperture at most,and the latter can reduce the system cost effectively.In the third part,to solve the strong mutual coupling effect resulting from the closely arrangement of array elements on the space-constrained carrier,a low computational complexity self-calibration two-dimensional direction finding algorithm for L-shaped array with unknown mutual coupling has been presented.It firstly transforms the data with unknown mutual coupling into the mutual compensated data by sacrificing some array elements data,which eliminates the influence of the mutual coupling,and then uses the propagator method to reduce the correlation effect between the source signals.Finally,with the complex eigenvalue technique,simultaneous estimation of two dimensional angles under unknown mutual coupling condition is achieved.Compared with the existing methods,the advantage of this method is that it can realize the two-dimensional direction finding in presence of the unknown mutual coupling with very low computational complexity and can deal with the highly correlated signals effectively,which is of great significance for transforming theoretical research on the L-shaped array direction finding system to engineering application.