| Large-scale MIMO antenna array whose elements amount to over one hundred,contains a wealth of space gain,which is mainly obtained through spatial multiplexing,transmission diversity and beamforming,bringing about high spectral efficiency and energy efficiency and making itself one of the key technologies of 5G.However,Massive MIMO systems have a large number of base station antennas,and intensive antenna deployment can result in that antenna subsets of high correlation coexist with antenna subsets of low correlation,furthermore,the correlation among the antennas will affect the system capacity.If the efficiency of antennas is not fully exploited according to the characteristics of the RF links,it will be difficult for the system to obtain the expected gain.Antenna grouping technique can divide the base station antennas into several groups using the correlation among antennas.Then,the high correlation antenna group will be used for beamforming and precoding design,eliminating the interference among users,while the low correlation antenna group can be used for spatial multiplexing,meeting the maximum data rate requirements,and ultimately make the system obtain higher spectral efficiency and better bit error rate performance simultaneously.However,the green communication vision requires future wireless network not only has high spectrum effect,but also should have high energy efficiency,so energy efficiency becomes critical for wireless communication network.Therefore,facing the characteristics of large-scale antennas array,this paper investigates the energy efficiency optimization problem in Massive MIMO system based on antenna grouping technique.The main contents are as follows:1)In this paper,we investigate antenna grouping for energy efficiency maximization in the downlink of single-user massive MIMO systems.We construct an antenna grouping system model which partitions large-scale transmit antenna array into fixed-size groups.System capacity,power consumption and energy efficiency are analyzed,based on the impact of the number of active transmit antennas per group.A low complexity algorithm for finding the optimal number of active transmit antennasin each group is proposed based on binary search.Simulation results show that when the channel correlation coefficient becomes larger,that is,the channel correlation becomes stronger,the antenna grouping can effectively improve the energy efficiency of the system,and there exists optimal value of the number of the active transmit antennas per group in the fixed-size grouping model that maximizes the energy efficiency.As an example at the transmitting power oftxP = 10 dBm,a 28%improvement in terms of energy efficiency is achieved by optimally assigning the number of antennas to each group,compared to the achievable performance with antennas deployed in the BS.2)In order to further improve the energy efficiency of Massive MIMO multi-user system based on antenna grouping,this paper proposes a joint optimization scheme of antenna grouping and power allocation.The scheme optimizes the power allocation and the optimal number of antennas for each group with maximum energy efficiency.Based on the proposed scheme,the mathematical model of energy efficiency optimization problem is constructed,which proves the existence of the optimal power distribution scheme.The optimization problem is solved by using the standard iterative optimization algorithm.Combined with the classical water-filling power allocation algorithm,the number of antennas and the power distribution between groups are combined to maximize the energy efficiency of the system.The corresponding solution is the optimal solution for the power allocation and the number of antennas in each group.The numerical simulation results show that the antenna grouping can improve the energy efficiency of the Massive MIMO system in the correlated channel,and the joint optimization of the antenna grouping and the power allocation technology can further improve the system energy efficiency. |
PDF Full Text Request