Efficient DOA Estimation Algorithm For Massive MIMO Systems

The fifth-generation mobile communication system is in the process of perfection and construction.Massive MIMO technology is considered to be one of the key technologies of 5G and has received extensive attention from the industry participant.In massive MIMO systems,efficient and accurate DOA estimation is a necessary prerequisite for the implementation of 3D-MIMO technology.Therefore,the two-dimensional DOA algorithm has extremely important significance for the entire massive MIMO system.At present,the super-resolution algorithm represented by the MUSIC algorithm and the ESPRIT algorithm is quite mature.However,in the massive MIMO system,the enormous complexity of the huge antenna array makes these algorithms no longer applicable.Therefore,an efficient DOA estimation algorithm has become an important research direction in Massive MIMO technology.Propagation operator algorithms(PM)have been widely used since they were proposed because of its low complexity.The algorithm uses linear operation instead of eigendecomposition of the entire received signal covariance matrix.The complexity of the algorithm is much lower than the traditional super-resolution algorithm.However,in the massive MIMO,the high complexity brought by the large-dimension matrix be operated and the angle matching problem brought by the two-dimensional DOA estimation still need further research and solution.This article focuses on efficient DOA estimation algorithms for Massive MIMO systems.For two-dimensional DOA estimation and algorithm complexity,an improved method based on PM algorithm is proposed.The main work of this article is summarized as follows:(1)The DOA estimation mathematical model and system assumptions are given.For the massive MIMO system,the planar rectangular array system model and the propagation operator algorithm based on the model are introduced in detail,and the algorithm complexity and estimation performance are analyzed.Aiming at the problem of parameter matching in multi-source 2D spatial DOA estimation,animproved parameter matching algorithm is proposed to suppress the angle blurring and has high robustness.(2)Continuing to solve the problems of 2D DOA estimation parameter matching and system model reconstruction,an automatic matching algorithm based on PM algorithm was proposed.A polynomial is constructed by the orthogonality of the noise subspace and the signal subspace.Parameters of the other dimension can be obtained by the estimated parameters and the 2D-parameters are automatically matched.The algorithm has obvious performance advantages at low SNR.(3)For Massive MIMO systems,the huge antenna array increases the dimension of the signal covariance matrix,and the complexity is pretty high.A cross-covariance Root-MUSIC PM algorithm was proposed.By dividing the antenna array into two sub-arrays,the mutual covariance matrix can be constructed,which reduces the dimensionality of the data matrix and reduces the computational complexity of the algorithm.At the same time,the cross-covariance matrix dimension can be adjusted.It can flexibly adjust the estimated performance and algorithm complexity.