Pilot Design And Optimization In Massive MIMO System

With the explosive growth of mobile service data,communication systems need to meet greater capacity and higher transmission rates.Because massive MIMO(Multiple input multiple output)systems can obtain greater gain and system capacity from the beamforming and spatial multiplexing of multiple antenna arrays,massive MIMO technology has become an important research direction fornext-generation wireless communications.Taking the massive MIMO system as an example,due to the rapid increase in the number of users in mobile communications,the number of pilots required for channel estimation in the system should also meet user needs.However,due to the limited number of pilots,this will inevitably cause the repeated use of pilots between different cells,resulting in pilot contamination(Pilot Contamination).Therefore,in a massive MIMO system,it is particularly important to establish a reasonable pilot allocation mechanism based on the inter-cell interference characteristics,user distribution,and cell characteristics,and to design the best pilot optimization algorithm to eliminate or reduce pilot pollution.Based on the massive MIMO system,this paper establishes a multi-cell multi-user system model,and designs a pilot allocation algorithm based on the interference between users.The main research work is as follows:(1)Based on the physical characteristics of wireless communication channel,the transmission characteristics of large-scale MIMO system are studied and analyzed,and the representation method of MIMO system channel in time domain is obtained.Based on the large-scale MIMO system model,the problem of pilot interference is studied,and the mathematical model of pilot pollution problem is established,which is the basis of pilot design and optimization.(2)By simulating the degree of interference between multiple cells and multiple users,the cell is divided into a central area and an outer area.The interference between users in the central area of this cell and the central areas of other cells is negligible,while the interference between users in the outer area and the outer area is very large,and an allocation algorithm is needed to reduce the impact of pilot pollution.In this paper,users in the central area can use the simplest method of randomly assigning pilots,which can significantly reduce the computational complexity without causing obvious pilot pollution.For users in the outer area,a weighted graph coloring algorithm is used to achieve pilot allocation.(3)In the large-scale MIMO system with multi cell and multi-user,the quantum bacteria optimization(qbfo)algorithm is combined with the pilot allocation scheme to search the optimal pilot allocation scheme by finding the optimal solution.Compared with the traditional bacterial colony algorithm and other optimization algorithms,the quantum bacterial colony algorithm has faster convergence speed and can search the optimal solution or near the optimal solution more quickly.The simulation results show that the qbfo algorithm can approach the optimal system rate in dealing with the pilot allocation problem,and the convergence speed of qbfo algorithm is faster than other swarm intelligence algorithms such as BFO and QPSO.