基于免疫粒群路径优化的网络拥塞控制研究

发布时间：2018-05-24 04:20

本文选题：人工免疫 + 粒群优化　；参考：《郑州大学》2014年硕士论文

【摘要】：随着近年来计算机应用的推广，人们对网络流量的需求以及网络规模的要求也越来越高，网络的飞速发展导致了时有网络拥塞现象的发生。为了保证网络的正常运行，就需要更高的网络服务质量（QoS）来满足网络发展的需求。网络拥塞的存在制约着网络的发展和应用，采取合理的措施来预防和控制网络拥塞的发生有着极其重要意义。用传统优化的方法来控制网络拥塞虽然在不断的完善，，但是总存在这样或者那样的问题。近年来，针对传统优化算法的不足之处，国内外的学者将智能优化算法应用到网络拥塞控制当中，通过智能优化算法来实现网络拥塞的控制成为了当前的一个研究热点。智能优化算法通过无意识的寻优行为来适应生存环境和优化生存状态的一种新型优化算法，到目前为止已经出现了多种智能优化算法。本文将人工免疫算法和粒子群优化算法结合在了一起应用到网络拥塞控制当中，具体的研究内容如下所述：（1）首先对网络拥塞现象、网络拥塞的成因、网络拥塞控制机制及其网络拥塞控制算法进行了分析，并对仿真软件NS2的应用进行了分析。（2）建立了研究网络拥塞控制的网络拓扑模型，并对网络拓扑模型和网络路由进行了分析和探讨。在对网络路径优化分析的基础上提出了网络拥塞路径优化问题。（3）将人工免疫和粒子群优化算法结合给出了一种免疫粒群优化算法。该算法将免疫算法中的免疫信息处理机制融合到粒子群优化算法当中，根据粒子群算法的收敛速度快的特点，利用人工免疫算法的特征多样性避免粒子群优化算法陷入局部解，提高了粒子群优化算法的后期收敛速度。（4）给出了一种基于免疫粒群优化的网络拥塞控制算法。以资源的消耗和负载均衡分布为网络路径优化目标，在满足带宽、时延、费用、等多项指标的前提下，使负载尽量均衡分布在有宽裕资源的链路上，提高了网络的吞吐量和传输效率，可有效的实现网络拥塞的优化控制。仿真结果表明了算法的有效性和可靠性。
[Abstract]:With the popularization of computer application in recent years, the demand for network traffic and network scale is becoming higher and higher. The rapid development of network has led to the phenomenon of network congestion. In order to ensure the normal operation of the network, a higher quality of service (QoS) is needed to meet the needs of network development. The existence of network congestion restricts the development and application of network. It is of great significance to take reasonable measures to prevent and control network congestion. Though the traditional optimization method is used to control the network congestion, there are always some problems. In recent years, in view of the shortcomings of traditional optimization algorithms, scholars at home and abroad apply intelligent optimization algorithm to network congestion control. The intelligent optimization algorithm to achieve network congestion control has become a hot research topic. Intelligent optimization algorithm is a new kind of optimization algorithm which adapts to the living environment and optimizes the living state by unconscious optimization behavior. Up to now, there have been many kinds of intelligent optimization algorithms. In this paper, the artificial immune algorithm and particle swarm optimization algorithm are combined in the network congestion control, the specific research content is as follows: Firstly, the phenomenon of network congestion, the cause of network congestion, the mechanism of network congestion control and the algorithm of network congestion control are analyzed, and the application of simulation software NS2 is analyzed. (2) the network topology model of network congestion control is established, and the network topology model and network routing are analyzed and discussed. Based on the analysis of network path optimization, the problem of network congestion path optimization is proposed. 3) an immune particle swarm optimization algorithm is proposed by combining artificial immune algorithm with particle swarm optimization algorithm. The immune information processing mechanism of the immune algorithm is integrated into the particle swarm optimization algorithm. According to the fast convergence speed of the particle swarm optimization algorithm, the artificial immune algorithm features diversity to avoid the particle swarm optimization algorithm into a local solution. The late convergence rate of PSO is improved. A network congestion control algorithm based on immune particle swarm optimization is proposed. Taking the resource consumption and load balance distribution as the network path optimization goal, under the premise of satisfying the bandwidth, delay, cost, and so on, the load can be distributed on the link with abundant resources as far as possible. The throughput and transmission efficiency of the network are improved, and the optimal control of network congestion can be realized effectively. Simulation results show that the algorithm is effective and reliable.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.06;TP18

【参考文献】