Research On Power Allocation For Downlink NOMA In Cellular Network

In the last fifty years,wireless communication technology has been exploited all aspects of life.With this comes the need for higher data rates and lower power consumption.The existing orthogonal multiple access(OMA)technology has been developed for a long time and becomes a mature technology,which also means that it is difficult to make a major breakthrough in performance.However,in non-orthogonal multiple access(NOMA)systems,multiple users can share the same time-frequency resources and cancel multiple access interference using superposition coding and serial interference cancellation,which significantly increases the maximum number of supported users and sum capacity.Power domain NOMA provides an additional dimension for designing multiple access,which attracts great attention due to its simple principle.In this thesis,the optimization design of power allocation algorithms for maximizing sum capacity and energy efficiency are studied.The thesis firstly takes the downlink single-carrier NOMA as an example to illustrate the principle and characteristics of NOMA,and compares it with traditional OMA systems.Then several basic user pairing algorithms and power allocation algorithms are introduced.At last,two power allocation algorithms for the downlink NOMA in cellular network are proposed.The first algorithm is proposed to find the globally optimal solution of the sum capacity maximization problem,which splits the joint power allocation into two steps: intrasubcarrier power allocation and inter-subcarrier power allocation.According to the capacity constraints of the problem,the optimal intra-subcarrier power allocation result can be calculated.And it is proved that the inter-subcarrier power allocation can be equivalently transformed to the power allocation for a virtual frequency division multiple access system,which can be solved by the proposed fast water-filling algorithm.In this part,the computational complexity analysis is also discussed.What’s more,the thesis proposes a lowcomplexity suboptimal user pairing algorithm.Then Monte Carlo simulations are used to evaluate the performance of the above algorithms,which proves that the sum capacity is significantly increased.The second algorithm is proposed to find the globally optimal solution of the energy efficiency maximization problem which is a nonlinear fractional programming problem.After some mathematical manipulations,a sufficient condition is proposed.When the condition is fulfilled,the EE maximization problem can be equivalently transformed to the SC maximization problem whose globally optimal solution is obtained by using the previous fast water-filling algorithm.When the condition is not fulfilled,the EE maximization problem is solved by using Dinkelbach algorithm with inner Lagrange dual with subgradient method or barrier algorithm with inner Newton’s method.Besides,the computational complexity analysis is also discussed in this part,which proves that the above algorithms can further reduce the dimension of the problem.At last,Monte Carlo simulations are used to evaluate the performance of the above algorithms.And it is proved that the energy efficiency is significantly increased.