Research Of The 3D Arrays DOA Estimation Based On The Nested Layout

DOA estimation is one of the applications of arrays signal processing in antenna arrays.Under the same DOA estimation algorithm,the DOA estimation error is affected by the number of the actual arrays elements(arrays aperture),signal-to-noise ratio and the number of snapshots.The increase of the number of arrays elements will undoubtedly lead to a sharp rise in the estimated cost.How to improve the accuracy of the DOA estimation is worth studying when the number of arrays elements is constant.In recent years,the proposal of the coprime arrays and the nested arrays has aroused people's widespread concern.The coprime arrays and the nested arrays can increase the degree of freedom of the arrays and expands the aperture of the arrays to improve estimation accuracy compared with the uniform linear arrays,and the coprime arrays and the nested arrays can detect more signal sources than the number of arrays elements.Based on the coprime arrays and the nested arrays,this paper mainly focuses on the estimation performance of the arrays in one-dimensional,two-dimensional and three-dimensional cases.The main research contents are as follows:Firstly,the DOA estimation of the one-dimensioanal arrays is studied based on the nested layout.The structures of the coprime arrays and the nested arrays are given.Virtual arrays elements are formed by the difference-position between the arrays elements can increase the degree of freedom of the arrays.The spatial smoothing MUSIC algorithm is applied to simulate the performance of the uniform linear arrays,the coprime arrays and the nested arrays.The simulation results show that the estimation performance of the nested array is the best,and the estimation performance of the uniform linear arrays is the worst.Moreover,the coprime arrays and the nested arrays can detect more spatial signal sources than the number of arrays elements.Secondly,the DOA estimation of the one-dimensioanal arrays is studied based on the nested layout.The two-dimensional(2D)coprime arrays model and the 2D nested arrays model are given based on the one-dimensional coprime arrays and nested array.The Cramer Rao Bound(CRLB)and the estimation error of the uniform linear area array,the coprime area array and the nested area array are simulated.The simulation results show that the nested area array has the lowest CRLB and the smallest error.The model of the L-shaped coprime arrays and the L-shaped nested arrays are given by the L-shaped arrays and simulating CRLB of the L-shaped linear arrays,the L-shaped coprime arrays and the L-shaped nested arrays.Based on the root MUSIC algorithm and compressed sensing algorithm,the estimation errors of the three categories L-shaped arrays are simulated with the signal-to-noise ratio,the number of snapshots and the number of actual arrays elements.Finally,we compare estimation error of the compressed sensing algorithm and the root MUSIC algorithm with the same parameters.Finally,we extend the 2D arrays to the three-dimensional(3D)case based on the nested layout and study the 3D area arrays and the 3D L-shaped arrays(2L-shaped arrays)based on the nested layout.The CRLB of the 3D arrays and the 3D L-shaped arrays are simulated,and the CRLB of the 3D area arrays and the 2D area arrays,the CRLB of the3 D L-shaped arrays and the 2D L-shaped arrays are compared with the same conditions.The results show that the CRLB of the 3D area arrays is lower than the CRLB of the 2D area arrays,and the CRLB of the 3D L-shaped arrays is lower than the CRLB of the 2D Lshaped arrays.This result means that the 3D arrays and the 3D L-shaped arrays are easier to accurately estimate the DOA of the spatial signal sources.For the 3D area arrays based on the nested layout,the Unitary-ESPRIT algorithm is applied to simulate the estimation error of the three categories 3D area arrays.For the 3D L-shaped arrays based on the nested layout,the root MUSIC algorithm and the compressed sensing algorithm are applied to simulate the estimation error of the three categories 3D L-shaped arrays.The results show that the estimation error based on the nested layout is the lowest.In addition,the CRLB of the area arrays and the 2L-shaped arrays are compared under the3 D condition,which proves that the 2L-shaped array is easier to accurately estimate the DOA of the spatial signal source.