双向反馈蚁群算法在网络负载均衡问题的研究  

Load balance and optimization of network resources based on twodirection feedback ant colony algorithm

作　　者：王爱静[1] 郝志峰[1] 黄翰[2,3] 李学强[2] 

机构地区：[1]广东工业大学计算机学院,广州510006 [2]华南理工大学软件学院,广州510006 [3]南京大学计算机软件新技术国家重点实验室,南京210093

出　　处：《计算机工程与应用》2011年第36期112-114,158,共4页Computer Engineering and Applications

基　　金：国家自然科学基金(the National Natural Science Foundation of China under Grant No.61003066;No.61070033);教育部博士点基金(No.20090172120035);中央科研业务费项目(No.2009ZM0052);广东省自然科学基金重点项目(No.9251009001000005);广东省科技计划项目(No.2010B050400011;No.2010B080701070;No.2008B080701005);广东省哲学社会科学规划"十一五"规划项目(No.08O-01);信息安全国家重点实验室开放课题基金(No.04-01)

摘　　要：针对网络资源管理中的负载均衡与优化问题,提出一种双向反馈蚁群算法,用蚂蚁数量代表网络资源流量,通过蚂蚁间信息素的相互作用和动态控制来实现网络流量分担到多条可用路径。将蚁群算法扩展为双向反馈的蚁群算法,蚂蚁判断各条路径上的信息素浓度的同时,考虑可选链路的负载情况,决定选择要走路径,使得蚂蚁相对均衡地分布在可选链路上。仿真实验结果表明,双向反馈蚁群算法比原蚁群算法在缩短自适应时间,减少丢包率,提高负载均衡效率方面都具有更好的性能。Two-direction feedback ant-colony algorithm is presented which aims at load balance and optimization of network resources management.Through the interaction and dynamic update among the pheromone of ants which are on behalf of the network traffic,the algorithm enables network traffic to share a number of paths available.The algorithm expands the ant colony algorithm for two-direction feedback ant colony algorithm.When the ant judges every path of pheromone strength,it also considers optional link load conditions,then determines which path to choice.It makes the link has the relatively balanced distribution.The results of simulation experiment demonstrate that compared with ant colony load balance algorithm two-direction feedback ant colony algorithm has superiority in reducing time of auto adaption,lowering packet loss rate and improving efficiency of load balance.

关 键 词：双向反馈蚁群算法 蚁群算法 网络资源优化 负载均衡 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]