Joint Optimization Of Spectral Efficiency And Energy Efficiency For Power-domain NOMA Systems

Emerging fifth generation mobile communications require data rates to increase by 100 to 1000 times and device connections by 10 to 100 times,placing higher demands on spectral efficiency(SE)and energy efficiency(EE).Since power-domain NonOrthogonal Multiple Access(NOMA)can greatly improve spectral efficiency,it has become a promising candidate for of the next generation mobile communications.However,explosive data growth has also brought huge energy consumption.Green communication is an important trend in future research.Based on the above research background,this paper first studies wireless resource allocation from the perspective of EE.However,it is difficult to fully consider the requirements of a communication system for single energy efficiency optimization.Therefore,in this paper,joint optimization of EE and SE is studied.In the EE study of the power-domain NOMA system,the mathematical model of EE optimization is given.In order to solve this hybrid optimization problem,it is decomposed into two sub-problems: user selection problem and power allocation.Considering the maximum transmission power of the base station,two low complexity user selection methods and a gradient descent based iterative power allocation scheme are proposed to maximize EE.And the impact of user selection on EE is studied.The simulation results show that compared to Orthogonal Frequency-division Multiple Access(OFDMA),the proposed resource allocation algorithm significantly improves EE performance.Considering the maximum transmission power and the minimum requirement rate,the multi-objective optimization of EE and SE for the power-domain NOMA system is studied.Firstly,the relationship between EE and SE for power-domain NOMA communication system is analyzed,and the closed-expression in the case of single carrier is theoretically derived.For SE and EE cannot be optimal at the same time,this paper proposes a multi-objective optimization framework.Using the weighted sum method,two competing performance metrics can be transformed into a single-objective optimization problem.According to different communication requirements,different weights can be determined to jointly optimize SE and EE.In this paper,it is proved that the performance of the power-domain NOMA system is related to the user channel gain.Therefore,a low complexity predefined user selection algorithm is proposed.When user selection is given,the power allocation problem is proved a quasi-convex problem.Therefore,an iterative two-layer power optimization method is proposed,where in the inner layer intra-subchannel power allocation is solved using Lagrangian multiplier method,and in the outer layer an iterative method based on the bisection method is proposed.Numerical simulation results show that the proposed resource allocation method can flexibly switch between SE optimization and EE optimization.The proposed algorithm has fast convergence speed and the EE performance is very close to the global optimal solution.Furthermore,the complexity of power allocation remains low with the increasing number of users.