Research On Multi-cell Cooperative Beamforming Algorithm

In the 5G mobile communication system,the macro base stations all use large-scale antenna arrays.When sending broadcast messages,narrow beams directed at different centers are used for scanning to complete data transmission.However,in the existing analog beamforming method,when the antenna arrangement is fixed,only narrow beams with a fixed width can be synthesized,and the beam shape is single,which sometimes causes insufficient coverage or large interference in neighboring cells.At this time,without changing the number and arrangement of antennas,a method of combining beams of different widths is urgently needed.In addition,in a certain area,how to reasonably plan the beams between base station sectors to increase system capacity while reducing neighboring cell interference?There is no more mature solution at present.For this reason,this paper uses an improved particle swarm algorithm to complete the construction of diversified beams with different widths and the joint tuning of multi-cell coordinated beamforming.This paper mainly includes the following four innovations:multi-channel pattern modeling method,intelligent construction of analog beams,5G system-level simulation platform construction,and improved particle swarm algorithm to achieve multi-cell cooperative beamforming.First,this article introduces the basic principles of analog beamforming in large-scale antenna systems,and proposes a superposition method for synthesizing beams with specific widths.In order to solve the shortcomings of the high weight loss of the superposition method,this paper adopts an improved particle swarm algorithm,namely the boundary adaptive particle swarm algorithm(BAPSO),which takes into account the various attributes of the composite beam and generates a rich and diverse codebook set which acts as candidates for the beam selection of the base station sector in each time slot.Next,the process of building a 5G system-level simulation platform is introduced.The main feature of the platform is to support the distribution of real base stations and the loading of features.Finally,based on the 5G system simulation platform,BAPSO is still used.After some adjustments,the multi-cell coordinated beamforming is completed by continuously tuning the beam selection of all base stations in the area.From the simulation results,when the BAPSO used in this paper generates a diversified codebook set,it has a faster convergence rate and is not easy to fall into a local area compared to the ordinary PSO algorithm,differential evolution algorithm and gray wolf optimization algorithm and other intelligent optimization algorithms.In the process of processing multi-cell cooperative beamforming,after comparative analysis of multiple sets of cases,the original BAPSO has been adjusted to make it the best optimization effect,and it is performed with random algorithms and expert experience algorithms under a variety of user distributions.Simulation and comparative analysis,among them,the adjusted BAPSO has a more prominent performance.In addition,the 5G system-level simulation platform used in this article has undergone a relatively complete system verification,which provides a reliable guarantee for algorithm research.Based on the above advantages,the multi-cell coordinated beamforming algorithm proposed in this paper is fully capable of landing in the existing network.