OFDMA系统资源优化综述

Review of resource optimization in OFDMA systems

在正交频分多址(OFDMA)系统中,有效的资源分配可以充分发挥系统性能.目前的资源优化主要分为频谱效率优化与系统能效优化两个方面.频谱效率优化即在给定的总功率下,最大化系统总容量;系统根据信道状态信息,分步对子载波和功率进行分配.根据不同的业务需求,频谱效率优化又细分为仅考虑频谱效率的优化以及考虑用户公平性与复杂通信环境下的优化.系统能效优化是在保证系统总容量的同时,降低系统的总功率消耗.能效优化算法采用了分式规划、约束条件松弛与拉格朗日对偶分解等最优化理论的方法进行资源分配的优化.在加入不同约束的考虑下,能效优化算法分为保证用户公平性、回程线路容量、通信安全等不同的能效优化方案.但是,这两方面的资源优化对不完整信道信息的情况考虑较少,对于多业务的资源分配,未能针对业务多样性进行高效资源优化.另外,资源优化在跨层资源分配上仍有不足.

In the Orthogonal Frequency Division Multiple Access （OFDMA） system, the effective resource allocation has apparent impact on the performance of the system. The current resource optimization mainly includes two aspects： spectral efficiency and energy efficiency. Spectral efficiency optimization maximizes the total capacity of the system under a given total power. According to the channel information, subcarriers and power is allocated by sequence, power allocation being conducted after subcarrier allocation. Depending on different business needs, there are the optimization that exclusively considers spectral efficiency, the optimization that considers user fairness and the optimization under complex environments. Energy-efficient optimization reduces the total system power, while ensuring the total system capacity. Energy-efficient optimization algorithm utilizes the optimization theory, such as fractional programming, constraint relaxation and Lagrangian dual decomposition. Under th condition of adding different constraints, there are energy-efficient optimization programs with user fairness guaranteed, and backhaul capacity guaranteed, and communication security guaranteed. But, these two aspects of resource optimization are still inadequate in considering incomplete channel information. For multi-business resource allocation, resource optimization cannot take advantage of business diversity. In addition, these two aspects of resource optimization are insufficient in cross-layer resource allocation.