中文摘要：

由于可以有效地提高频谱效率,能量效率与前程效率,云接入网络（C-RAN）被认为是未来第五代无线网络中的重要组成部分。不同于传统蜂窝网络,在云接入网络中,基带处理单元（BBU）被从基站分离,并聚合成一个中央计算云。无论如何,这些优化目标（频谱效率,能量效率,前程效率）在大多数情况下相互冲突,并且单个目标性能提升通常会导致其他目标性能的下降。据作者所知,在云接入网络中的多目标优化（MOO）问题,仍未被考虑过。在本文中,我们针对基于正交频分多址（OFDMA）的云接入网络,设计对应的联合优化算法以解决多目标优化问题。仿真结果显示,比起仅考虑单目标优化,本文提出的算法可以有效的解决不同优化目标之间的权衡,并且为云接入网络的资源分配提供一个新的方向。

英文摘要：

The Cloud access network（ C-RAN） is considered as an important part of the future 5G wireless network,because it can effectively improve the spectrum efficiency（ SE）,energy efficiency（ EE） and fronthaul efficiency（ FE）. In most cases,these optimization goals（ SE,EE,and FE） conflict with each other,and the performance improvement of one target usually leads to other targets performance degradation. As far as the author is concerned,the multi-objective optimization（ MOO） in the cloud access network has not been considered. In this paper,we intend to design joint optimization algorithms to solve the multi-objective optimization problem in orthogonal frequency division multiple access（ OFDMA） based cloud access networks. In the special case of single-user and single-RRH,we use a Lagrange dual decomposition method to solve the non-linear fractional programming problem and we design a Modified Particle Swarm Optimization（ M-PSO） algorithm to solve the more complex issue for the general case of multiple users and multi RRHs. The simulation results show that the algorithms proposed in this paper can effectively solve the trade-off among different objectives compared with singleobjective optimization and provide a new direction for the resource allocation of cloud access network.