基于多代理的认知卫星通信架构及频谱感知算法的研究

发布时间：2018-01-14 00:22

本文关键词：基于多代理的认知卫星通信架构及频谱感知算法的研究　出处：《北京交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：无线电是卫星通信网络的主要通信手段之一,目前主要采用时分复用技术,但是该技术未能有效地利用现有频谱资源,通信容量的扩充也只是依靠高频带宽来完成。因此,如何在卫星通信中充分利用有限的频谱资源提高网络吞吐量,避免出现频谱资源的"假性枯竭"问题,是目前研究的热点问题之一。本文首先对卫星通信网络以及认知无线电(Cognitive Radio,CR)基础知识进行了介绍,从理论的角度分析CR技术在卫星通信网络的可行性和先进性。在此基础上,重点研究了基于多代理的认知卫星通信架构以及基于干扰的频谱感知算法。论文主要工作如下:(1)提出了基于多频谱感知代理(Spectrum-sensing Agent,SA)的认知卫星通信架构。首先,分析了认知卫星动态接入空闲授权频段的可行模式。针对传统CR网络在卫星通信中面临的挑战,提出由SA代替认知用户执行频谱感知,并建立了基于多SAs的认知卫星通信架构。然后具体介绍了该架构的网络框架、功能框架以及频谱感知运行模式和帧结构。最后提出以通信吞吐量作为性能指标。该架构不仅降低了认知用户硬件设计的复杂度、成本以及额外的能耗,还增加了认知用户的通信时间,进而提高了认知网络的通信吞吐量。(2)提出了基于干扰的多代理频谱感知算法,并得到不同情况下认知用户以及网络的平均可实现吞吐量。该算法根据垂直模式下认知卫星利用地面授权频段做上行通信的情况,建立详细的网络模型和多代理频谱感知模型。利用主用户接收机的信噪比门限定义干扰概率,根据距离分布函数得到整个网络的干扰概率期望。通过干扰建模,分析认知用户和主用户共享相同频段的可能性,最后得到不同情况下认知用户以及网络的平均可实现吞吐量。(3)仿真对比实验验证了基于干扰的多代理频谱感知算法的性能。针对认知卫星利用地面授权频段做上行通信的情况,使用不同融合规则和主用户信噪比门限,从吞吐量方面将基于干扰的多代理频谱感知算法与传统CR算法进行比较。结果表明,该算法的平均吞吐量不再随着感知时间的增加而减少,并且相较于传统CR算法,提高了认知用户以及网络平均可实现吞吐量,有利于进一步实现频谱资源利用率的提高。本文最后进行了总结,并对认知卫星通信后续在宽带频谱感知技术、融合决策算法以及数据库资源动态管理和更新等方面的研究进行了展望。
[Abstract]:Radio is one of the main means of communication in satellite communication networks. At present, time division multiplexing (TDM) technology is mainly used, but this technology can not make effective use of the existing spectrum resources. The expansion of communication capacity only depends on high frequency bandwidth. Therefore, how to make full use of limited spectrum resources to improve network throughput in satellite communications, and avoid the problem of "pseudo exhaustion" of spectrum resources. This paper first introduces the basic knowledge of satellite communication network and Cognitive Radio CRs. The feasibility and advancement of CR technology in satellite communication network are analyzed theoretically. The architecture of cognitive satellite communication based on multi-agent and spectrum sensing algorithm based on interference are studied in detail. Spectrum-sensing Agent. Firstly, the feasible mode of cognitive satellite dynamic access to idle authorization band is analyzed. Aiming at the challenges of traditional CR network in satellite communication. A new cognitive satellite communication architecture based on multiple SAs is proposed, which is implemented by SA instead of cognitive users. Then, the network framework of this architecture is introduced in detail. Finally, the communication throughput is proposed as the performance index. This architecture not only reduces the complexity, cost and additional energy consumption of the hardware design of cognitive users. In addition, the communication time of cognitive users is increased, and the communication throughput of cognitive networks is improved. (2) A multi-agent spectrum sensing algorithm based on interference is proposed. The average throughput of cognitive users and networks under different conditions is obtained. The algorithm makes uplink communication according to the vertical mode of cognitive satellites using terrestrial authorized frequency bands. A detailed network model and a multi-agent spectrum sensing model are established. The interference probability is defined by the signal-to-noise ratio threshold of the primary user receiver, and the interference probability expectation of the whole network is obtained according to the distance distribution function. Analyze the possibility of cognitive users and primary users sharing the same frequency band. Finally, the average realizable throughput of cognitive users and networks under different conditions is obtained. The simulation results verify the performance of the multi-agent spectrum sensing algorithm based on interference, aiming at the uplink communication of cognitive satellite using the ground authorized frequency band. Using different fusion rules and signal-to-noise ratio (SNR) threshold of the primary user, the interference based multi-agent spectrum sensing algorithm is compared with the traditional CR algorithm in terms of throughput. The average throughput of the algorithm is no longer reduced with the increase of perceptual time, and compared with the traditional CR algorithm, the average throughput of cognitive users and network can be improved. It is helpful to further improve the efficiency of spectrum resources. Finally, this paper summarizes the cognitive satellite communication follow-up in broadband spectrum sensing technology. The research of fusion decision algorithm and database resource dynamic management and update are prospected.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN925;TN927.2

【参考文献】