【作者】 张婷；

【导师】 郑凤；

【作者基本信息】 北京邮电大学 ， 电子与通信工程（专业学位）， 2021， 硕士

【摘要】 由于无线通信设备的增长和网络技术的发展,对频率的需求不断增长。为了在有限的频率资源环境中有效地共享频率,应该进行研究以开发频谱共享技术。传统的频谱共享研究依靠中央机构来验证每个频谱共享交易的真实性,缺少安全的频谱共享机制,这容易受到众多的安全威胁。其次,通过频谱感知,或频谱数据库进行的传统频谱共享机制使用效率并不是很高。最后,由于同频道干扰和其他干扰,自私且理性的频谱所有者不愿在没有适当经济补偿的情况下共享频谱资源。在未来的5G范式中,将部署数十亿台移动设备,以实现广域和无处不在的数据感测,收集和传输。考虑到物联网通信的流量特性和频谱短缺的困境,解决传统频谱共享时依靠中央机构来认证和验证每个频谱共享交易,且缺乏安全的频谱共享激励机制问题,论文使用区块链的分布式记账功能来进行频谱管理。区块链特有的功能,不仅可以确保访问和数据传输的安全性,而且通过智能合约编写应用程序进行交换数据可以保证交易过程的有效性。因此,本文研究了基于区块链的频谱共享的关键技术,提出了一种区块链网络中基于博弈论的动态频谱共享激励算法,在此之上,设计并实现了基于联盟链的动态频谱交易系统。(1)在5G异构网络中,将有各种设备连接到互联网中,实现物联网(Internet of Things,IoT)和万物互联(Internet of Everything,IoE)。针对各种移动设备频谱资源利用率不足、运营商即主用户(Primary Users,PUs)之间缺乏激励策略的问题,论文研究了区块链互联网网络中多个PUs之间与移动设备即次用户(Second Users,SUs)共存的主用户的频谱定价问题,提出了 一种由伯川德(Bertrand)博弈设计的频谱共享算法,以分析多个PUs间的频谱价格博弈过程。首先,设计频谱共享网络模型和交易模型,研究主系统之间的竞价博弈,并构建Bertrand博弈模型,PUs作为博弈竞争者,目的是最大化运营商利润。其次,通过设计合理的效用函数,将Bertrand竞争模型证明为超模博弈,进而证明纳什均衡解的存在性。然后,根据SUs的频谱需求,将博弈模型分为静态博弈和动态博弈,分别进行纳什均衡的求解。最后,通过增加一个自适应变异机制以改进粒子群算法对纳什均衡的最优快速寻解策略,博弈论模型的仿真结果表明:论文提出的算法经过迭代最终收敛于纳什均衡时,价格是PUs效用最大的策略。(2)在激励算法的设计与研究的基础上,解决了传统频谱共享网络中缺乏安全性,频谱交易过程不透明,使用权得不到保障的问题。在需求分析及Fabric项目的体系架构和核心模块的基础上,设计了基于联盟链的频谱共享系统架构。将频谱交易过程,频谱管理和交易清单列表写入频谱总账构成频谱交易平台,形成分布式频谱状态数据库。此外,在频谱交易规则中,用智能合约链码代替合约,实现链上支付,对交易数据的监控和可追溯,形成完整的频谱共享流程,以实现频谱资源的高效共享和安全控制,提高频谱利用率。更多还原

【Abstract】 Due to the growth of wireless communication equipment and the development of network technology,the demand for frequency continues to grow.In order to effectively share frequencies in a limited frequency resource environment,research should be conducted to develop spectrum sharing technologies.The traditional spectrum sharing research relies on the central authority to verify the authenticity of each spectrum sharing transaction.The lack of a secure spectrum sharing mechanism makes it vulnerable to numerous security threats.Secondly,the efficiency of the traditional spectrum sharing mechanism through spectrum sensing or spectrum database is not very high.Finally,due to co-channel interference and other interference,selfish and rational spectrum owners are unwilling to share spectrum resources without proper financial compensation.In order to solve the problem of the traditional spectrum sharing relying on the central agency to authenticate and verify each spectrum sharing transaction,and the lack of a secure spectrum sharing incentive mechanism,the paper uses the blockchain’s distributed accounting function technology for spectrum management.The unique functions of the blockchain can not only ensure the security of access and data transmission,but also can ensure the effectiveness of the transaction process by writing applications to exchange data through smart contracts.Therefore,this paper studies the key technology of spectrum sharing based on blockchain,and proposes a dynamic spectrum sharing incentive algorithm based on game theory in blockchain network.On top of this,the dynamic spectrum trading system based on alliance chain is designed and implemented.(1)In the 5G heterogeneous network,various devices will be connected to the Internet to realize the Internet of Things(IoT)and the Internet of Everything(IoE).Aiming at the problems of insufficient utilization of spectrum resources and lack of incentive strategies among primary users(PUs).The thesis studies the spectrum pricing of primary users coexisting with secondary users(SUs)among multiple PUs in the blockchain Internet network.A spectrum sharing algorithm designed by Bertrand game is proposed.The purpose is to analyze the spectrum price game process among multiple PUs.Firstly,the spectrum sharing network model and transaction model are designed to study the bidding game between the main systems.Construct a Bertrand game model in which PUs act as game competitors.The purpose of this model is to maximize its profit.Secondly,by designing a reasonable utility function the Bertrand competition model is proved to be a supermodel game.And then prove the existence of Nash equilibrium solution.Then,according to the spectrum requirements of SUs the game model is divided into static game and dynamic game and the Nash equilibrium is solved respectively.Finally,an adaptive mutation mechanism is added to improve the particle swarm algorithm’s optimal fast-searching strategy for Nash equilibrium.The simulation results of the game theory model show that when the algorithm proposed in the paper finally converges to the Nash equilibrium after iteration.The price is the most effective strategy for PUs.(2)On the basis of the design and research of the incentive algorithm,the problems of the lack of security in traditional spectrum sharing networks and the opaque spectrum trading process and the inability to guarantee the right of use are solved.Based on demand analysis and the architecture and core modules of the Fabric project.A spectrum sharing system architecture based on the alliance chain is designed.The spectrum trading process,spectrum management and transaction list are written into the spectrum general ledger to form a spectrum trading platform and form a distributed spectrum status database.In addition,in the spectrum trading rules,smart contract chain codes are used instead of contracts to realize on-chain payment,monitoring and traceability of transaction data and forming a complete spectrum sharing process.In turn,efficient sharing and security control of spectrum resources are realized and spectrum utilization is improved.更多还原