| With the rapid development of research on intelligent driving technology,the research of advanced driving assistance systems has become a research hotspot.Vehicle-mounted millimeter-wave radar has gradually become an important part of the sensor system of the intelligent driving system because of its small size,low cost,and the ability to adapt to harsh environments freely.In the research of vehicle-mounted millimeter-wave radar,the direction of arrival(DOA)estimation is an important part of the research of vehicle-mounted millimeter-wave radar target positioning,and it is an important performance index of vehicle-mounted millimeter-wave radar target positioning.This thesis analyzes and studies the DOA estimation problem based on MIMO(Multiple input multiple output)arrays.Aiming at the failure of traditional DOA methods when multiple sources are used,a new method of MIMO radar-based DOA estimation(ICA-CS)is proposed.;In addition,to solve the problem of DOA estimation performance degradation when coherent sources appear,a two-dimensional DOA estimation(MSR-ESPRIT-like)method for MIMO radar based on Toeplitz matrix set reconstruction is proposed.The main research contents of this thesis are as follows:1.Introduced the current research status of vehicular millimeter-wave MIMO radar at home and abroad,and summarized the principle of FM continuous wave speed measurement and ranging.The focus is on the DOA estimation methods that affect the performance of MIMO radar.The Capon algorithm,MUSIC algorithm,ESPRIT algorithm,compressed sensing theory and other algorithms are studied,and the above algorithms are simulated and analyzed,and the advantages and disadvantages of each algorithm are analyzed and summarized.2.Aiming at the problem that the traditional DOA method cannot estimate the multi-source signal,the coprime array model is studied,and a new parallel MIMO coprime array model is proposed,which can obtain more virtual arrays with fewer actual elements.The number of array elements,the use of a coprime array to improve the degree of freedom of the array,and the combination with the compressed sensing theory,the use of the sparsity of the array to improve the accuracy of DOA estimation.The improved DOA estimation method(ICA-CS)in this thesis is compared with the traditional compressed sensing DOA estimation and Bayesian compressed sensing DOA estimation methods.Experiments show that this method can effectively distinguish a large number of signal sources,and at the same time,it has better performance.High accuracy of angle estimation.3.Aiming at the problem of DOA estimation performance degradation when the signal is a coherent source,it is proposed to divide the two-dimensional array into two one-dimensional arrays for two-dimensional DOA estimation,construct the Toplitz matrix set of the signal matrix,and use the signal and The relationship between noise reconstructs the signal and performs related processing,and uses forward and backward translation to further improve the accuracy,and uses the ESPRIT algorithm for DOA estimation.The Toeplitz matrix reconstruction method(MSR-ESPRIT-like)in this article is combined with ESPRIT-like and ESC-FBSS-ESPRIT method is compared with simulations,and it is verified that the method can effectively reduce the influence of coherent source signals on DOA estimation,and at the same time has high angle estimation accuracy.In general,the research in this article focuses on the problem of radar DOA estimation methods,and solves the problem of the failure of the traditional DOA estimation method of multi-sources and the performance degradation of the traditional DOA estimation method under coherent sources from different perspectives,which makes up for the traditional DOA estimation method of MIMO radar.The lack of estimation accuracy provides theoretical support for the development of high-precision positioning technology for vehicle-mounted MIMO radar. |
