Research On Energy Efficiency And Spectral Efficiency Trade-off Optimization In Massive MIMO Systems

The mobile internet technology continues to thrive,which drives the access of a massive number of terminals.Although wireless communication technologies have enabled high data transmission,they may also lead to an urgent matter of energy crunch.Since increasing attentions are paid to the environment conservation and energy consumption,information transmission with high energy efficiency(EE)has become a destination for future communication systems to pursue,especially in 5G systems which declare to improve the EE by a factor of 100.Hence,EE has emerged as an important metric in green communication system design.Many investigations provide insight into the optimization of energy efficiency at present,while in fact,there is a conflicting association between the energy efficiency and spectral efficiency(SE).Therefore,it is significant to investigate the trade-off optimization between EE and SE.Massive MIMO system relies on the spatial multiplexing gain offered by a large number of antennas,and has been shown to potentially improve system capacity without increasing the bandwidth.Meanwhile,extra antennas help the user terminals to achieve high data rates using low-cost and low-precision components.Thus,massive MIMO is an efficient approach to improve the energy and spectral efficiency.However,the implement of massive MIMO requires a dedicated ratio chains for each antenna,which brings a surge of energy consumption and might degrade the energy efficiency.In order to maximize the energy efficiency and spectral efficiency in massive MIMO systems,the resource allocation and the optimization of hybrid precoding architecture are studied in this paper as follows:Firstly,the transmit power and the number of antennas at the base station are optimized in massive MIMO systems.The multipath channel and energy consumption in massive MIMO system are modeled,on the basis of which the impact of transmit power and the number of antennas on the EE and SE is analyzed,and then the EE-SE trade-off relationship is expounded.Due to multi-objectivity and multi-parameters,the EE-SE trade-off optimization can be solved with multi-objective optimization algorithms.In this paper,a multi-objective adaptive genetic algorithm(MAGA)is proposed with novel evolutionary mechanism and simple non-dominated sorting method.The simulation results have shown that,compared with other multi-objective optimization algorithms,MAGA can approximate the perfect Pareto front of EE-SE trade-off with fast convergence.As a result,the optimal configuration scheme of transmit power and the number of antennas can be obtained under different design requirements.Secondly,to deal with the heavy cost and power consumption issues caused by the large number of radio frequency(RF)chains in massive MIMO system,the hybrid analog and digital precoding is effective with relatively less RF chains.In this paper,the delay lines are found to be substitutes for phase shifters in analog precoding.They are not only low-power but able to break through the constant modulus constraints of phase shifters,thus optimal unconstrained hybrid precoding algorithm can be used to maximum SE.Based on the delay lines,the hybrid precoding architectures with partially connected RF and fully connected RF are considered respectively,and the corresponding curves of EE-SE trade-off optimization are simulated.Compared with the phase shifters networks,the hybrid precoding with delay lines has shown potential advantages in improving the EE-SE performance.What's more,the fully connected architecture seems more applicable to gain maximum EE for high SE designs.