Research On Antenna Beamforming In 3D MIMO Scene In 5G Network

With the increasing innovation of mobile communication technology,the human will enter the new era of communication,the fifth generation mobile communication(5G).In the 5G era,large-scale 3D MIMO and beamforming are key technologies in 5G networks,which can effectively increase spectrum and energy efficiency and reduce inter-cell interference.Since 3D MIMO adds vertical dimensional spatial components,combined with large-scale antenna arrays,beam pointing resolution is greatly improved,but it also increases the difficulty of beamforming.Therefore,the research of beamforming algorithms in large-scale 3D MIMO systems is of great significance.After analyzing the principle of 3D MIMO technology,Antenna array structure,array receiving signal model and the common beamforming algorithm,this paper studies the beamforming algorithm in large-scale 3D MIMO scene.The specific work is as follows:First,for the case where the beamforming weights in large-scale 3D MIMO systems are difficult to obtain the optimal value,the intelligent particle swarm optimization algorithm is used to solve the weights based on the maximum signal-to-noise ratio criterion.The computational complexity of obtaining better beam weights.Secondly,for large-scale 3D MIMO systems,the standard particle swarm optimization algorithm is used to solve the problem that the weights of the particles are premature and the convergence speed is not good.The standard particle swarm optimization algorithm is improved: First,chaotic initialization is used to make the initial particles.The distribution is more reasonable,and the chaotic perturbation of the particle position beyond the spatial range in each iteration effectively avoids the situation that the particle stagnate at the local extremum for a long time and is difficult to jump out;secondly,comprehensively considers the iteration number and the particle performance versus the inertia factor Influence,different inertia factors are adopted for different performance particles,so that the particles can reduce the search for poor regions and search for better regions more fully;furthermore,introduce dynamic neighborhood operators and make full use of the sharing mechanism of particle swarm optimization algorithm.Further increase the diversity of the particles.Thirdly,for the problem of steering vector error existing in the actual beamforming process,the improved particle swarm optimization algorithm is used to correct the mismatch factor,which minimizes the mismatch factor;and on the basis of the guidance mismatch correction The flap constraint further reduces sidelobe interference.Finally,the proposed algorithm is simulated.The results show that the proposed algorithm not only can form a beam with a more accurate azimuth,but also achieves the mismatch correction and sidelobe interference,and achieves the beam pattern.