中文摘要：

为分享在属于多操作员的收音机存取网络(RANs ) 之间的光谱的讨价还价的基于的机制被学习，到改进光谱利用效率并且最大化网络收入。由介绍一个聪明的代理人，各跑了能力，它包括交换信息交换的、最后的决策，等等与另外的 RANs 交换光谱。分享机制的建议内部操作员的光谱与不完全的信息，并且结果讨价还价作为一场无限地平线的讨价还价比赛被建模比赛有唯一的顺序的平衡。因而，实现基于分析被精制。模拟结果证明建议机制超过常规固定光谱管理(FSM ) 在网络收入，光谱效率，和调用堵住率的方法。

英文摘要：

Bargaining based mechanism for sharing spectrum between radio access networks （RANs） belonging to multioperators is studied, to improve spectrum utilization efficiency and maximize network revenue. By introducing an intelligent agent, each RAN has the ability, which includes trading information exchanging, final decision making, and so on, to trade the spectrum with other RANs. The proposed inter-operator spectrum sharing mechanism is modeled as an infinite-horizon bargaining game with incomplete information, and the resulting bargaining game has unique sequential equilibrium. Consequently, the implementation is refined based on the analysis. Simulation results show that the proposed mechanism outperforms the conventional fixed spectrum management （FSM） method in network revenue, spectrum efficiency, and call blocking rate.