Research On Two-Dimensional Angle Estimation Algorithm Based On The Distributed Nested Array

As a key technology in array signal processing,direction of arrival(DOA)estimation has important applications in radar,sonar,communication and other systems.Two dimensional(2-D)DOA estimation can more accurately describe the spatial feature information of incident signal and obtain more accurate signal source location,so it has attracted extensive attention of scholars.Compared with uniform array,non-uniform sparse array has more advantages The advantages are: 1.The array aperture is larger;2.More sources can be estimated;3.Better direction-finding(DF)resolution;4.Higher DF accuracy.This makes it possible to improve the performance of two-dimensional DOA estimation.Among the commonly used 2-D DOA estimation arrays,the advantage of L-shaped array is that the array structure is more practical and convenient for analysis and processing.Therefore,aiming at the problem of 2-D DOA estimation of L-shaped array,the following research is carried out in this paper:Firstly,the 2-D DOA estimation technology based on L-shaped nested array is studied.In order to obtain a continuous virtual array with significantly increased degrees of freedom,which can be constructed to realize the underdetermined estimation of source’s DOA with more than the number of array elements after vectorization of the received data covariance matrix.Based on the different construction methods of received data covariance matrix,two kind of 2-D DOA estimation algorithms based on L-shaped array are studied.The first is to construct Toeplitz matrix after vectorization for the cross covariance matrix and its conjugate matrix between different subarrays,which is solved by ESPRIT algorithm;The second is to obtain the covariance matrix of pseudo data based on conjugate augmented spatial temporal(CAST)technology and space-time received data model according to the spatial and time-domain characteristics of the received signal.The first one is solved directly by unitary joint diagonalization algorithm without spatial smoothing,and the second one is solved by unitary ESPRIT algorithm after 2-D spatial smoothing.The simulation results show that the2-D DOA estimation based on space-time received data model not only reduces the computational complexity,but also has the best estimation performance.Secondly,aiming at the problem of limited array aperture expansion of L-shaped nested array,the distributed array structure is introduced.The nested array is regarded as a subarray of distributed array.In order to improve the array freedom and reduce the mutual coupling effect,an array structure of L-shaped distributed extended nested array is proposed.Aiming at the problems of angle pairing,angle ambiguity and limited estimation ability in DOA estimation of L-shaped sparse array,combined with the process of structuring space-time received data model,a new 2-D DOA estimation algorithm based on 2-D vectorization conjugate enhancement processing and two-scale unitary ESPRIT is proposed.Simulation results show that this method greatly improves the performance and resolving power of DOA estimation on the basis of persisting the goodness of nested array and distributed array.Finally,the proposed new array structure is applied to the bistatic MIMO radar system for joint estimation of 2-D DOD and 2-D DOA,and an new array structure of L-shaped distributed augmented nested array based on the bistatic MIMO radar is proposed.Combined with the two-scale ESPRIT algorithm,it not only completes the high-precision estimation of the four-dimensional angle,but also realizes the 2-D angle pairing of the transceiver array.Simulation results show that this method can effectively improve the resolution and angle measurement accuracy of the system in angle estimation,and the pairing algorithm is easy to implement.