Research On Technologies For Mitigating Pilot Contamination In Multi-cell Massive MIMO Systems

With the increasing demands for higher data-rate transmission in mobile communication,Massive MIMO becomes one of the key technologies for the fifth generation of wireless communications(5G)because of its outstanding ability in improving energy and spectral efficiency.As a consequence,multi-cell Massive MIMO networks are the most promising networks for 5G.The system performance of multi-cell Massive MIMO networks is connected to the number of orthogonal pilot sequences which are used for estimating the channel state information(CSI).The number of orthogonal pilot sequences is restricted by the coherence time and it cannot satisfy the demands of huge number of users,which causes the serious inter-cell interference called pilot contamination.However,the effect of pilot contamination cannot disappear when the number of antennas at BS goes to infinity and pilot contamination is the bottleneck problem for improving the performance of Massive MIMO systems.So,in this paper,to reduce the effect of pilot contamination,we focus on two techniques--pilot allocation and pilot contamination precoding(PCP).The main work and the innovations in this paper can be summarized as follows:(1)A pilot allocation scheme is proposed which can provide not only a good overall performance but also the fairness among users.The existing pilot allocation schemes can be divided into two types roughly.One is to maximize the sum-rate of the system and the other one is to maximize the minimum rate in the system.Both of the two types have disadvantages:The one maximizing the sum-rate of the system can get a good overall performance but ignore the fairness among users in the system and the one maximizing the minimum rate in the system can guarantee the fairness among users but reduce the sum-rate largely.Almost no pilot allocation work can guarantee the two aspects at the same time.To solve the problem,in this work,we proposed an optimization problem based on the criterion maximizing the product rate of all users in the system,which can provide not only a good overall performance but also the fairness among users.To solve the proposed optimization problem,the User Exchange Based on Greedy algorithm(UEBG)and the Min Leakage algorithm are developed.Simulation results show that the pilot allocation schemes proposed in this paper can provide not only a good overall performance but also the fairness among users with low computational complexity.(2)A PCP scheme which can get a good overall performance is proposed considering finite number of antennas at BSs.In practice,the number of antennas cannot be arbitrarily large because of the limitation of space and the complexity of antenna devices.For practical consideration,this paper proposes a PCP scheme assisted sum-rate maximization for multi-cell massive MIMO systems with finite antennas at BSs.When the number of antennas at BSs is finite,the impact of noise,channel estimation error,channel non-orthogonality and channel uncertainty caused by the usage of statistical CSI cannot be ignored compared with pilot contamination.We in this work jointly consider all these factors and formulate an optimization problem to maximize the sum-rate of all users.The proposed scheme can provide a good overall performance with acceptable computational complexity,especially for the case where users suffer from strong pilot interference.