Research On Pilot Allocation In Massive Distributed Antenna Systems

With the rapid development of mobile communication business and the arrival of big data era,people put forward more and higher requirements for mobile communication technology.As a new network architecture,large-scale distributed antenna system(DAS)can not only obtain the spatial diversity gain of traditional MIMO system,but also effectively integrate many key physical layer technologies in the new generation of mobile communication technology(5g),including network coverage technology,spectrum expansion technology,etc.At the same time,because the antenna elements of the distributed antenna system are more dispersed in geographical location,shorter transmission distance and higher energy efficiency can be achieved.In the distributed system,in order to make full use of the spatial diversity gain of MIMO system,it is also necessary to estimate the channel.However,due to the limited time and frequency resources of the channel,the orthogonal pilot sequences that the system can use are also limited.Users assigned to the same pilot will interfere with each other,that is,pilot contamination,which will reduce the system capacity and increase the bit error rate.Thesis mainly studies how to minimize the influence of pilot contamination on the system performance through pilot allocation in large-scale distributed antenna system.The main work of thesis includes the following aspects.Firstly,it introduces the main characteristics of wireless channel and analyzes the common models of wireless channel: path loss model,large-scale fading model and small-scale fading model.Because channel estimation is mainly to use precoding technology to combat channel fading and improve system capacity,the paper then introduces the main precoding technology system,and focuses on two commonly used precoding technologies: MMSE precoding and block diagonalization precoding.Then it introduces the channel estimation technology,mainly about MMSE estimation.Secondly,the influence of pilot contamination on system performance is analyzed,and the application of heuristic algorithm in large-scale distributed MIMO system is studied.Two algorithms are studied: greedy algorithm and tabu search algorithm.The idea of greedy algorithm is relatively simple,but the performance is not very ideal,and the users allocated last will get worse performance,it is not fair to those users,and the computational complexity is also higher.The tabu search algorithm is a heuristic algorithm,which starts from an initial allocation scheme and realizes the effective search of solution space by constantly searching to the neighborhood and allowing the appearance of bad solution to a certain extent.Compared with greedy algorithm,the performance of tabu search algorithm is improved to some extent,but the computational complexity is still not ideal.Finally,by analyzing the optimal allocation scheme obtained by exhaustive algorithm in different scenarios,it is found that there is a certain relationship between the advantages and disadvantages of pilot allocation scheme and the distance between users,and it is theoretically proved that the farther the distance between users,the smaller the pilot interference problem will be.Based on this,a pilot assignment algorithm based on the distance between users in large-scale distributed MIMO system is proposed.The simulation results show that the proposed algorithm based on the distance between users is very close to the optimal allocation scheme obtained by the exhaustive algorithm in both estimation error and system and rate,but the complexity is still high.In order to reduce the complexity,another pilot allocation algorithm based on user exchange is proposed in thesis,which can approximate the performance of ergodic optimal pilot allocation scheme with linear complexity.