【作者】 王娜娜；

【导师】 孔金生；

【作者基本信息】 郑州大学 ， 控制理论与控制工程， 2007， 硕士

【摘要】 近年来，随着网络规模和网络流量的飞速增长，网络上越来越多的是具有QoS要求的多媒体业务。但是，传统的路由算法往往约束单一又不能充分利用现有的网络资源，经常导致有些链路被过度使用，有些链路却空闲不用的负载不均衡局面，从而致使网络拥塞，业务的服务质量无法得到保证。网络拥塞已成为制约网络发展和应用的瓶颈。在这种情况下，如何在满足QoS要求的前提下，通过路径优化来实现网络拥塞控制是一个崭新的课题。本文对基于遗传粒群路径优化的网络拥塞控制方法进行了研究，其主要工作和内容如下：(1)在对网络拥塞、拥塞控制、拥塞预防及网络路由对拥塞的影响进行分析的基础上，给出了利用网络路径优化解决网络拥塞控制问题的思想。对网络仿真软件NS2进行功能扩展，并将其应用于网络仿真中，取得了较好的效果。(2)对QoS及QoS路由进行了详细分析，在对网络拥塞分析的基础上，对网络拥塞路径优化进行了深入地探讨，为实现网络拥塞提供了条件。(3)将粒群(PSO)和遗传算法(CA)相结合，给出了遗传粒群优化算法。算法初期经过了前端的PSO优化处理，在初始种群里就有很大的概率包含最优解。随着进化代数的不断增加，该算法能够快速找到最优解，而在算法后期采用遗传算法引入新的个体，可以避免算法过早陷入局部最优解。从而使算法在速度和精度上同时得到提高。将其用于解决多峰值函数优化问题中，仿真表明了该优化算法的有效性和可靠性。(4)提出了基于遗传粒群路径优化的网络拥塞控制方法，该方法在满足带宽、延迟、费用多项QoS指标的条件下对负载进行路径优化，以负载均衡分布函数和资源消耗函数作为优化目标，旨在消耗尽可能少的网络资源的同时，也使网络负载的分布尽量均衡，从而避免拥塞。给出了算法实现及其仿真分析，仿真结果表明该算法的有效性和可靠性。更多还原

【Abstract】 In recent years, with the rapid growth of the network scale and traffic, there are more and more multimedia applications. They require the network to provide quality of service(QoS) guarantees. But traditional routing algorithms, with single constrain condition, cannot make full resource utilization, which always lead to the unbalanced traffic distribution, some links getting over-utilized, while others remaining under-utilized. It results in congestion and the quality of services not guaranteed. Network congestion has become a bottleneck which restricts the growth and application of the network. In this case, on the basis of satisfying QoS requests, how to realize network congestion control by optimizing network path is a brand-new research subject.This thesis mainly focuses on the research of network congestion control method based on GAPSO Path Optimization Algorithm. The main contents are as follows:(1) Based on the analysis of network congestion control, network routing and quality of service, an idea is proposed to solve the network congestion control problem by using network path optimization. To extend the function of NS2 and apply it into network simulation can get a better result.(2) QoS and QoS routing is analyzed in detail. On basis of analysis of network congestion control, a discussion is made in the paper to achieve network congestion.(3) A GAPSO algorithm is presented by combining Particle Swarm Optimization Algorithm and Genetic Algorithm. In initial phase, PSO Algorithm is adopted to get a new population. Then, it makes use of the GA Algorithm to optimize the result. With the evolution generations increase, the proposed algorithm can get the best value rapidly and meanwhile it can avoid converging at local best value. Also, it improves in computing speed and precision. Applying it in the problem of multi-peak function Optimization, simulation results demonstrate the validity and feasibility of the proposed algorithm.(4) A network congestion control method is put forward based on GAPSO path optimization. The method optimizes the path of network load and makes network load balancing and resource consumption as aim function to avoid congestion by balancing the load and minimizing network resource consumption in the condition of meeting bandwidth, delay and cost constrains. Algorithm realization and simulation are presented in the paper. The simulation results show that the algorithm is effective and reliable.更多还原