Research On Technologies For Mitigating Pilot Contamination In Massive MIMO Systems

Massive Multiple-Input Multiple-Output(MIMO)technology,using base stations equipped with hundreds of antennas to serve multiple users in a cell,can effectively enhance the spectral efficiency and energy efficiency of wireless communication systems.However,because of the shortage of both time and frequency resources,it is inevitable to reuse pilot sequences in massive MIMO systems,which leads to interference between users with identical pilot in different cells.This problem is known as pilot contamination.The performance of massive MIMO systems is limited by the impact of pilot contamination due to inter-cell interference.Pilot assignment schemes,reasonably arranging users assigned the same pilot in the system,can effectively reduce interference between them.This thesis focuses on the pilot contamination problem of TDD massive MIMO systems and mainly study how to mitigate the impact of pilot contamination by assigning pilots to users properly.The detailed contents are as follows:First of all,this thesis studies the pilot contamination problem of massive MIMO systems.Starting from the development of MIMO technology,the massive MIMO technology is introduced.After that,the channel models,channel estimations and the data transmission process of massive MIMO systems are analyzed.On this basis,the reason and the impact of the pilot contamination problem are discussed.Then,the location-based pilot assignment algorithm is taken into consideration.In the conventional pilot reused schemes,there is a difference in the fairness between users in the system and the accuracy of channel estimation is lower than expected.In this thesis,a improved location-based pilot assignment scheme is proposed,which assigns pilot sequences in turn from the user that is the farthest from the base station in its cell.The pilot sequence assigned to a user is selected when the sum of potential mutual interference between users in the system that reuse the same pilot is lowest.Simulation results show that the proposed scheme provides a good channel estimation accuracy and increases the minimum uplink achievable rate of the system.Last,the vertex graph coloring-based pilot assignment algorithm is studied.When the number of antennas at base station is limited,an interference graph of the massive MIMO system is constructed by analyzing the potential mutual interference of two users.Based on the above,a vertex graph coloring-based pilot assignment algorithm is proposed.The criterion for selecting one pilot for a user in the proposed algorithm is that the newly added interference is minimal,thus reducing the overall interference of the whole system.The simulation results show the significant improvement of the proposed algorithm compared with existing methods in the performance of the massive MIMO system.