Research On Target Direction Estimation For MIMO Radar With Sparse Array

Multiple-Input Multiple-Output(MIMO)radar employs multiple transmit and receive elements,and has the ability to jointly plan transmissions and process received signals.It has been the focus of research after it was proposed.In this paper,we concentrate on investigating the angle estimation of CMIMO radar with sparse array.First,sparse array optimization problem of MIMO radar is studied.Then the problem of angle estimation of MIMO radar with sparse array is discussed.The main work of this paper is listed as follows.1.Sparse array optimization problem of MIMO radar is studied.In order to obtain better transmit and reception pattern,a modified Discrete Artificial Bee Colony based on Cloud model is proposed for optimizing MIMO radar sparse array.The roulette wheel selection scheme employed by onlookers choosing the good-quality nectars is.the key to fast convergence of Artificial Bee Colony.However,the selection scheme is so greedy that it also leads to the rapid decrease of population diversity and fall into local optimization.So we use Cloud model as the new selection scheme to improve Artificial Bee Colony.Because randomness and stable tendency of Cloud model could maintain the diversity of population and fast convergence.And it not only guarantees the superiority of excellent individuals,but also pays attention to the effect of inferior individuals.However,the local search is introduced to improve search accuracy.Simulation results illustrate the proposed method has excellent performance.Ensuring the mainlobe width in a certain value,it could obtain lower sidelobes.2.The problem of angle estimation of MIMO radar with sparse linear array is discussed.Sparse linear arrays can extend the array aperture and spatial resolution as compared to uniform linear arrays with the same number elements.In this paper,we introduce the model of sparse linear array into MIMO radar to improve the accuracy of angle estimation.But,the disadvantage of sparse linear array is manifold ambiguity which can cause large angle estimation errors.In order to avoid manifold ambiguity in MIMO radar with sparse linear array,a new ESPRIT-based estimator is proposed based on multi-scale method.The main idea is based on a combination of two estimates to solve the manifold ambiguity.First,we use ESPRIT algorithm to obtain ambiguous estimates,then the unambiguous estimates is obtained by selection matrix.And the unambiguous estimates serves as coarse references to disambiguate the ambiguous estimates.Then the unambiguous and high solution angle estimation is obtained.However,when the element of sparse linear array is too many,the dimension of selection matrix increases sharply.So we propose the ESPRIT-MUSIC algorithm to avoid the manifold ambiguity.It also obtains ambiguous estimates by ESPRIT algorithm.And the ambiguous candidate estimates is obtained by ambiguous estimates.Then we use MUSIC-based procedure to search the true angle estimate from a set of ambiguous candidate estimates to obtain accurate angle estimation.Simulation results illustrate ESPRIT-MUSIC algorithm could improve the accuracy of angle estimation and has excellent performance.But,the requirement of two-dimension search renders much higher computational cost.In order to reduce the complexity for angle estimation,we derive a Reduced-dimension ESPRIT-MUSIC algorithm.It simplifies MUSIC-based procedure by two times one-dimension.And it could obtain high solution angle estimation.Simulation results show that it reduces the algorithm complexity and ensures the performance of angle estimation.3.Problem of the angle estimation of MIMO radar with arbitrary sparse array is studied.As sparse linear array,arbitrary sparse array could extend the array aperture.And the position of sparse array elements has little constraint.But,the disadvantage of sparse array is manifold ambiguity which can cause false angle estimation.In order to avoid manifold ambiguity in MIMO radar with sparse array,a new solve ambiguity method is propose based on MUSIC algorithm.The main idea of the proposed method is obtained transmit angles and receive angles separately by Reduce-MUSIC algorithm.When transmitter exists the ambiguity,we compare the MUSIC spectrum of orignal transmit array with the MUSIC spectrum of sliding transmit array to obatin the real transmit angle estimation which could coincide in the transmit MUSIC spectrum.Then we obtain the real receive angle estimation through the real transmit angle estimation.Similar,When receiver exists the ambiguity,we compare the MUSIC spectrum of orignal receive array with the MUSIC spectrum of sliding receive array to obatin the real receive angle estimation which could coincide in the receive MUSIC spectrum.Then we obtain the real transmit angle estimation through the real receive angle estimation.And the transmit angle and the receive angle could auto-matching.So it could realize cross location to targets.