认知无线电系统中一种改进的功率控制博弈算法  

An improved power control game in cognitive radio system

作　　者：康钰[1,2] 朱琦[1,3] 酆广增[1] 

机构地区：[1]南京邮电大学江苏省无线通信重点实验室,江苏南京210003 [2]重庆邮电大学通信与信息工程学院,重庆400065 [3]东南大学移动通信国家重点实验室,江苏南京210096

出　　处：《电路与系统学报》2010年第2期52-58,共7页Journal of Circuits and Systems

基　　金：国家重点基础研究发展计划资助(2007CB310607);国家自然科学基金(60772062);东南大学移动通信国家重点实验室开放研究基金资助课题(N200813)

摘　　要：分布式功率控制是认知无线电(CR)系统中的关键技术之一,它直接影响到无线系统的性能。本文采用了博弈论的方法来实现对CR用户的分布式功率控制,在David Goodman的非协作博弈算法的基础上,给出了一种改进的效用函数,它使各用户在满足要求信干比条件下发送功率最小,同时使整个系统内由各种干扰引起的失真最小。本文通过理论推导证明了新的效用函数存在纳什均衡,并且均衡点唯一,同时仿真验证了该算法的收敛性,仿真结果表明这种兼顾用户自身利益以及用户间公平性的效用函数能降低发送功率,并且有效提高CR系统的性能。Distributed power control is one of the key techniques in cognitive radio system, since it directly affect the performance of wireless system. The transmit-power control problems in a cognitive-radio environment （involving multiple users） may be viewed as a game-theoretic problem. Based on the non-cooperative power control game, which is created by David Goodman, an improved utility function is given. This function can minimize the average transmit-power and interference under the condition of satisfying a required SIR. It is proved that the Nash Equilibrium is existed and unique for the improved game. Meanwhile, the simulation also certificates the convergence of the new algorithm. Simulation results show that the new algorithm which gives attention to the balance between efficiency and fairness can reduce transmission power efficiently and the performance of cognitive radio system is improved significantly.

关 键 词：认知无线电 功率控制 博弈论 效用函数 公平性 

分 类 号：TN929.5[电子电信—通信与信息系统]