基于博弈论和免疫克隆算法的认知无线网络资源的优化分配

发布时间：2018-02-28 13:23

本文关键词： 认知无线电 OFDM 频谱分配博弈论免疫克隆算法　出处：《西安电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：在无线电领域中,认知无线电技术的应用极大地减少了传统静态频谱分配的资源浪费,提高了频谱利用率,改善了整个通信系统的性能,在此基础上采用合理的策略优化信道通信性能是一个极具挑战性的课题。认知无线网络中的资源主要包括频谱带宽、发射功率、调制方式、频谱感知方法及感知时间等,这些资源的合理分配影响着整个无线网络的信道容量及系统Qo S(Quality of Service,服务质量),因此这些资源的优化分配成为提高通信性能的关键技术。认知OFDM网络系统是将OFDM(Orthogonal frequency-division multiplexing)技术应用于认知无线电中,这种多载波调制技术降低了各信道之间的相互干扰,提高了信号传输质量,其中的资源主要包括子载波和发射功率等,如何最优化地分配这些资源成为近年来的研究热点。基于此,本文主要研究在认知无线电的频谱共享接入模式下提高频谱利用率,以及对认知OFDM网络中的子载波和功率进行优化分配提高系统信号传输质量和通信性能。本文的主要工作如下:1.基于认知无线网络中合作用户的频谱共享模式,提出一种基于博弈论模型的免疫克隆优化算法。首先,选择潜博弈对频谱建立优化分配的模型,根据计算机网络中的TCP/IP协议中的三次握手机制,对频谱进行初步分配,经过有限次的博弈得到纳什均衡点。然后,使用免疫克隆算法优化该纳什均衡点,通过克隆、变异、选择等操作,寻找最优解。最终,本文算法提高了用户的SIR(Signal to Interference Ratio,信干比)及系统信道容量,得到了较好的信道分配性能。2.基于博弈论思想,利用完全信息动态博弈对认知OFDM中的子载波和功率分配问题进行建模,将每个子载波看作博弈的参与者,将所有用户看作每个博弈者的选择策略,考虑系统的总信道容量、功率约束及比例公平约束等条件,合理设置效益函数,最终通过有限次的博弈寻找最优解,增大了系统吞吐量。3.研究认知OFDM网络系统的下行链路子载波和功率的优化分配,提出一种基于免疫克隆选择算法的改进算法。本文选择两步式的分配方法:(1)将子载波分配给用户;(2)将功率分配给子载波。首先使用近似贪婪的比例公平初始化方法生成初始子载波种群,初步实现用户吞吐量分配的比例公平性。然后利用罚函数法将系统总的吞吐量、功率约束及比例公平约束条件等转化为评价函数对子载波和功率进行优化分配,寻找适应度最高的抗体,解码即为子载波和功率的最优分配方案。最终,系统的整体通信性能得到较大的提高,用户吞吐量比例公平性也接近理想。
[Abstract]:In the field of radio, the application of cognitive radio technology has greatly reduced the waste of resources in traditional static spectrum allocation, improved the spectrum efficiency and improved the performance of the whole communication system. On this basis, it is a challenging task to optimize the channel communication performance by adopting reasonable strategies. The resources in cognitive wireless networks include spectrum bandwidth, transmit power, modulation mode, spectrum sensing method and sensing time, etc. The rational allocation of these resources affects the channel capacity of the whole wireless network, the quality of service and the quality of service of the system. Therefore, the optimal allocation of these resources becomes the key technology to improve the communication performance. Cognitive OFDM network system is to make OFDM(Orthogonal. Frequency-division multiplexing technology is used in cognitive radio, This kind of multicarrier modulation technology reduces the mutual interference between different channels and improves the quality of signal transmission. The resources mainly include subcarriers and transmit power, etc. How to optimize the allocation of these resources has become a hot topic in recent years. Based on this, this paper mainly studies how to improve the spectrum efficiency under the shared access mode of cognitive radio. As well as optimizing the subcarrier and power allocation in cognitive OFDM network to improve the signal transmission quality and communication performance. The main work of this paper is as follows: 1. Based on the spectrum sharing mode of cooperative users in cognitive wireless networks, An immune clone optimization algorithm based on game theory model is proposed. Firstly, the optimal allocation model of spectrum is established by using latent game, and the spectrum is initially allocated according to the three-order handshake mechanism of TCP/IP protocol in computer network. After a finite number of games, the Nash equilibrium point is obtained. Then, the immune clone algorithm is used to optimize the Nash equilibrium point, and the optimal solution is found through cloning, mutation, selection and so on. The proposed algorithm improves the SIR(Signal to Interference ratio (signal-to-interference ratio) and the channel capacity of the system, and obtains a better channel allocation performance. 2. Based on the game theory, Using the complete information dynamic game to model the subcarrier and power allocation problem in the cognitive OFDM, each sub-carrier is regarded as the participant of the game, and all users are considered as the selection strategy of each gamer, and the total channel capacity of the system is considered. Under the condition of power constraint and proportional fairness constraint, the benefit function is set reasonably, and the optimal solution is found through a finite game, which increases the throughput of the system. 3. The optimal allocation of subcarrier and power in the downlink of cognitive OFDM network system is studied. This paper proposes an improved algorithm based on immune clone selection algorithm. In this paper, we choose a two-step allocation method: 1) assign subcarriers to users / 2) and allocate power to subcarriers. First, we use an approximate greedy proportional fair initializer. Method to generate the initial subcarrier population, Then the penalty function is used to convert the total throughput, power constraints and proportional fairness constraints of the system into evaluation functions to optimize the allocation of subcarriers and power. In order to find the antibody with the highest fitness, decoding is the optimal subcarrier and power allocation scheme. Finally, the overall communication performance of the system is greatly improved, and the fairness of the user throughput is close to the ideal.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN925

【相似文献】