摘要
针对主次用户博弈中信息为不完全信息的情况,提出了一种基于动态贝叶斯博弈的主用户频谱策略。在给定目标信噪比的情况下,将主用户的发射功率看作是其出租频谱代价函数的影响因素之一。根据发射功率的不同,得出主用户的两种不同的代价类型,再根据海萨尼转换,将主用户的多种类型组合,最后利用动态博弈方法分析主用户的频谱策略和收益。仿真结果表明,若博弈的某一方认为其它博弈方出现高噪声的概率逐渐增加,该博弈方在保证自身通信质量的前提下将相应增加其出租的频谱数量,从而获得更大收益。
To combat the problem of uncomplete information in the game for primary and secondary users,this paper proposes a spectrum scheme based on dynamic bayesian game theory.Given the target SNR,the emmitting power of the primary user becomes a factor of the cost function.According to the different emmitting power level,two cost functions are derived.Then the types of every primary user are combined using the Harsanyi transformation.The spectrum strategy and the revenue of primary users are analyzed using dynamic Bayesian Game Theory under every possible set.Simulation results show that when one Game player thinks the possibility of the other player under high noise is ascending,it will acquire more revenue by increasing the spectrum leasing in terms of guaranteeing its communication quality.
出处
《南京邮电大学学报（自然科学版）》
北大核心
2013年第5期1-6,共6页
Journal of Nanjing University of Posts and Telecommunications：Natural Science Edition
基金
国家科技重大专项(2011ZX03005-004-03)
江苏省高校自然科学研究基金(12KJB510014)
江苏高校优势学科建设工程资助项目
关键词
认知网络
贝叶斯博弈
频谱策略
代价函数
cognitive radio networks（CRN）
Bayesian game
spectrum strategy
cost function
