Resource Allocation Of Multi-Cell Multi-Carrier Based 5G System

Resource allocation is the critical part to guarantee the efficient operation,which decides the performance of communication systems and the quality of service of users.Most existing works on resource allocation focus on the case of single cell and it is not sufficient for the more complicated situation of multiple cells.And many researches considered the traditional orthogonal frequency division multiplexing(OFDM)in 4G as the modulation scheme.However,OFDM is sensitive to carrier frequency offset and introduces significant inter-carrier interference.Other multi-carrier schemes such as filter bank multi-carrier(FBMC),universal filtered multi-carrier(UFMC)and generalized frequency division multiplexing(GFDM)are seldom considered in resource allocation.Aiming at the characteristic of dense coverage in 5G networks,the key of this work lies in the resource allocation of multi-cell multi-carrier based 5G networks.According to the three main performance indicators: spectrum efficiency,energy efficiency and joint efficiency,the research contents and contributions can be concluded as follows.(1)To maximize the capacity of multi-cell with the limited spectral resource,an algorithm based on multiple access channel and non-cooperative game is proposed,and the performances of spectral efficiency for different modulation schemes are evaluated.Due to the binary variables,multiple access channel is utilized to remove these variables in objective and constraints;then the power allocation problem is formulated as a non-cooperative game and it is proven that the Nash equilibrium can be achieved;at last,the gradient projection method is applied to solve the strategy of each cell.The experimental results indicate that the proposed algorithm converges rapidly and can realize the high utilization of spectrum efficiency.(2)To maximize the energy efficiency with reducing power consumption and increasing capacity of multi-cell,an algorithm based on alternate optimization and successive convex approximation is proposed,and the performances of energy efficiency for different modulation schemes are assessed.In this algorithm,a heuristic scheme based on maximal signal interference and noise ratio is adopted to remove the binary variables;and an equivalent transformation is used to eliminate the ratio structure;the non-convex problem is transformed into a sequential convex problem through geometric mean approximation at last.The experimental results show that the proposed algorithm converges and achieves higher energy efficiency compared to some existing algorithms.(3)For optimizing the joint efficiency with the incompatibility of spectrum and energy efficiency,a tradeoff metric framework is proposed and successive convex approximation is employed to solve the optimization problem of the framework.Firstly,the joint efficiency is expressed as a term of multi-objective;then by introducing the exponent weights to each objective,a single objective can be constructed,and a specific character is achieved from the theoretical proof;finally,the successive convex approximation is sought for the optimization of single objective.Simulation results demonstrate that the proposed algorithm can solve the joint efficiency problem of multi-cell.All of the proposed algorithms can be directly applied to multi-cell communication systems.The multi-carrier performance comparisons will provide some relevant references for the 5G networks even the 6G networks.