一种改进的多约束混沌选播路由算法  

An Improved Anycast Routing Algorithm on Chaos with Multiple Constraints

作　　者：史蕊[1] 袁彩虹[2] 

机构地区：[1]河南大学软件学院,河南开封475004 [2]河南大学计算机与信息工程学院,河南开封475004

出　　处：《河南大学学报（自然科学版）》2014年第4期479-483,共5页Journal of Henan University:Natural Science

基　　金：国家自然科学基金资助项目(61004006);河南省科技厅资助项目(132300410149)

摘　　要：为满足带宽、时延、时延抖动、包丢失率等多服务质量约束下以最小代价进行选播路由,首先分析了多约束选播路由算法难以收敛于全局的原因,然后提出了一种改进的混沌选播路由算法.算法引入相异度的思想,把相异率作为判断"早熟"的方法;设计了混沌扰动算子,在判断种群出现"早熟"时对种群进行混沌扰动,增加群体的多样性,扩大搜索范围,避免陷入局部最优.该选播路由算法能找到满足多QoS约束的全局最优解.通过对比仿真实验数据表明,本算法具有较快的收敛速度和较高的请求成功率.In order to achieve the goal of least-cost anycast routing with multiple Quality-of-Services （QoS） constraints, such as bandwidth, delay, delay-jitter and packet loss ratio. Based on studying the reason of difficulty to falling into global optimum solution in multiple anycast routing, an improved anycast routing algorithm based on chaos genetic algorithm is proposed. The dissimilarity idea is introduced that can use the different rate as a way to judge the premature convergence. A chaos disturbance operator is designed to disturb the populations when judges the populations appear premature, so to increase the diversity of population, expand the scope of the search and avoid falling into local optimal. The algorithm can effectively find the global-best solution that satisfied the QoS constrains. The comparison of the experimental results further demonstrates that the proposed algorithm can accelerate the convergence rate and improve the success rate of finding the optimal solutions.

关 键 词：选播 混沌 相异度 扰动算子 遗传算法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]