潜博弈在认知无线网络中的信道选择应用研究

发布时间：2018-03-24 14:03

本文选题：认知无线网络　切入点：潜博弈　出处：《哈尔滨工业大学》2017年硕士论文

【摘要】：近年来,传统无线网络技术正逐渐难以满足爆发式增长的通信需求。而以频谱为代表的无线类资源正变得越来越稀缺紧张,认知无线网络技术正逐渐受到人们的普遍关注。博弈论作为研究具有竞争现象的理论和方法,已经在解决认知无线网络中的资源分配相关问题中发挥着重要作用。潜博弈作为一种特殊类型的博弈,相比较于其他类型博弈,其具备有限递增属性、纳什均衡的存在性和唯一性等诸多良好属性。本课题主要利用潜博弈理论解决认知无线网络技术中的与信道选择相关的两个关键问题——频谱分配和广播重传。频谱分配作为认知无线网络中的关键技术,已知的潜博弈模型在设计效用函数和潜在函数时只考虑了用户之间相互干扰这一主要因素。本文在已有模型的基础上进行改进,综合考虑了干扰量和吞吐量这两个因素来设计新的效用函数和潜在函数,并给出了新的潜博弈模型PG-TI。在PG-TI基础上设计出基于新的潜博弈模型的频谱分配算法PG-TIA。仿真实验结果表明,本文提出的PG-TIA算法具有良好的收敛性,而且能够有效增大系统吞吐量、降低系统总的传输功率,并表现出良好的算法公平性。广播重传是认知无线网络数据包广播过程中必须要考虑到的问题,其目的是为了实现可靠的数据传输。而广播重传发生时,用户如何选择合适的信道进行重传操作决定着重传的效率和质量。本文设计了一种基于潜博弈的广播重传模型PG-BR。在该模型中,设计了两种广播重传的协议机制来避免“冲突”问题,并针对不同申请使用信道的用户情况设定了优先级。在效用函数和潜在函数设计中综合考虑了能耗和干扰两大因素。并提出了一种新的基于潜博弈的广播重传算法PG-BRA。仿真实验结果表明,本文提出的PG-BRA算法能够增大数据包可达率、减少数据包平均重传次数、降低平均时延和降低系统总的传输功率。
[Abstract]:In recent years, the traditional wireless network technology is gradually unable to meet the explosive growth of communication needs. The wireless resources, represented by the spectrum, are becoming increasingly scarce. Cognitive wireless network technology is getting more and more attention. Game theory is a theory and method to study the phenomenon of competition. The latent game, as a special type of game, has the attribute of limited increment, compared with other types of game, which plays an important role in solving the problem of resource allocation in cognitive wireless network. The existence and uniqueness of Nash equalization and many other good properties. This paper mainly uses the latent game theory to solve the two key problems related to channel selection in cognitive wireless network technology: spectrum allocation and broadcast retransmission. Spectrum assignment is a key technology in cognitive wireless networks. In the design of utility function and potential function, the known latent game model only takes into account the main factor of user interference. A new utility function and a potential function are designed by considering both interference and throughput factors. A new latent game model PG-TI. a spectrum allocation algorithm PG-TIA-based on the basis of PG-TI is designed. The simulation results show that the proposed PG-TIA algorithm has good convergence and can effectively increase the system throughput. Reducing the total transmission power of the system and showing good algorithm fairness. Broadcast retransmission is a problem that must be considered in the process of packet broadcast in cognitive wireless networks. The aim is to achieve reliable data transmission. And when broadcast retransmission occurs, In this paper, a broadcast retransmission model PG-BRbased on latent game is designed. Two protocol mechanisms for broadcast retransmission are designed to avoid "conflict". In addition, the priority is set for users applying for different channels. Two factors, energy consumption and interference, are considered in the design of utility function and potential function. A new broadcast retransmission algorithm based on latent game is proposed. Method PG-BRA.The simulation results show that, The proposed PG-BRA algorithm can increase the reachability of data packets, reduce the average number of retransmissions, reduce the average delay and reduce the total transmission power of the system.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN92

【参考文献】