Opportunistic Spectrum Access in Unknown Dynamic Environment: A Game-Theoretic Stochastic Learning Solution

Opportunistic Spectrum Access in Unknown Dynamic Environment: A Game-Theoretic Stochastic Learning Solution
复制标题

未知动态环境中的机会频谱访问:博弈论随机学习解决方案

DOI:
10.1109/twc.2012.020812.110025
复制
发表时间:
2012-02
影响因子:
10.4
通讯作者:
Qihui Wu
Qihui Wu
中科院分区:
计算机科学1区
文献类型:
--
作者:
Yuhua Xu;Jinlong Wang;Qihui Wu

文献摘要

参考文献

被引文献

相似文献

我们研究了分布式信道选择问题,在机会频谱接入(OSA)系统中使用博弈论的随机学习解决方案的信道可用性统计和次级用户的数量是先验未知的。我们制定的渠道选择问题作为一个游戏,这被证明是一个确切的潜在的游戏。然而,由于缺乏其他用户的信息和限制,频谱是时变的未知的可用性统计,实现纳什均衡(NE)点的游戏是具有挑战性的任务。首先,我们提出了一个基因辅助算法来获得NE点的假设下,完善的环境知识。在此基础上,我们调查的系统吞吐量和公平性方面的游戏可实现的性能。然后,我们提出了一种基于随机学习自动机(SLA)的信道选择算法,次级用户从他们个人的行动奖励历史中学习,并调整他们的行为向NE点。该学习算法既不需要信息交换,也不需要关于信道可用性统计和次级用户数量的先验信息。仿真结果表明,基于SLA的学习算法在保证系统公平性的同时,还能获得较高的系统吞吐量。
We investigate the problem of distributed channel selection using a game-theoretic stochastic learning solution in an opportunistic spectrum access (OSA) system where the channel availability statistics and the number of the secondary users are apriori unknown. We formulate the channel selection problem as a game which is proved to be an exact potential game. However, due to the lack of information about other users and the restriction that the spectrum is time-varying with unknown availability statistics, the task of achieving Nash equilibrium (NE) points of the game is challenging. Firstly, we propose a genie-aided algorithm to achieve the NE points under the assumption of perfect environment knowledge. Based on this, we investigate the achievable performance of the game in terms of system throughput and fairness. Then, we propose a stochastic learning automata (SLA) based channel selection algorithm, with which the secondary users learn from their individual action-reward history and adjust their behaviors towards a NE point. The proposed learning algorithm neither requires information exchange, nor needs prior information about the channel availability statistics and the number of secondary users. Simulation results show that the SLA based learning algorithm achieves high system throughput with good fairness.
DOI: --
发表时间: 1998-09
期刊: ArXiv
影响因子: --
作者:
R. Jain;D. Chiu;W. Hawe
通讯作者: R. Jain;D. Chiu;W. Hawe
DOI: 10.1109/lsp.2010.2047315
发表时间: 2010-04
影响因子: 3.9
作者:
Wei Zhong;Youyun Xu;M. Tao;Yueming Cai
通讯作者: Wei Zhong;Youyun Xu;M. Tao;Yueming Cai
DOI: 10.1109/jsac.2004.839380
发表时间: 2005-02-01
影响因子: 16.4
作者:
Haykin, S
通讯作者: Haykin, S
DOI: 10.1109/98.788210
发表时间: 1999-08-01
期刊: IEEE PERSONAL COMMUNICATIONS
影响因子: --
作者:
Mitola, J;Maguire, GQ
通讯作者: Maguire, GQ
DOI: 10.1109/tpds.2007.70814
发表时间: 2008-11
影响因子: 5.3
作者:
Tarun Joshi;Disha Ahuja;Damanjit Singh;D. Agrawal
通讯作者: Tarun Joshi;Disha Ahuja;Damanjit Singh;D. Agrawal