软件定义卫星光网络蚁群优化波长路由技术  

Ant colony optimization routing and wavelength technology for software-defined satellite optical networks

作　　者：石晓东 李勇军[1] 赵尚弘[1] 王蔚龙 Shi Xiaodong;Li Yongjun;Zhao Shanghong;Wang Weilong(Information and Navigation College,Air Force Engineering University,Xi'an 710077,China)

机构地区：[1]空军工程大学信息与导航学院,陕西西安710077

出　　处：《红外与激光工程》2020年第10期203-210,共8页Infrared and Laser Engineering

基　　金：国家自然科学基金(91638101,61701522)。

摘　　要：针对卫星光网络中网络拓扑动态时变和业务类型多样化的问题,研究了在软件定义网络架构下保障服务质量的路由技术,提出了一种基于多业务的卫星光网络蚁群优化波长路由算法。通过改进蚁群算法的启发函数,将波长空闲率、时延、时延抖动、丢包率作为蚂蚁选路的重要依据,为业务选择了满足多种服务质量的最优路径;采用分组波长分配方法对不同等级的业务进行了区分服务,为不同业务分配了不同的波长集。仿真结果表明:与CL-ACRWA算法和Dijkstra算法相比,降低了卫星光网络的平均时延、平均时延抖动、平均丢包率,提高了波长利用率,同时也降低了高优先级业务的网络拥塞概率。Aiming at the problem of dynamic time-varying network topology and diversified service types in satellite optical network, the routing technology was studied to guarantee quality of service(QoS) under the framework of software-defined network(SDN), and a multi-service based ant colony optimization wavelength routing algorithm for satellite optical network was proposed. The heuristic function of ant colony algorithm was improved. The wavelength idle rate, time delay, time delay jitter and packet loss rate were taken as the important basis of ant routing, and the optimal path satisfying various QoS was selected for business. The group wavelength assignment strategy(GWAS) was used to distinguish services of different levels. Different sets of wavelengths were assigned to different services. The simulation results show that compared with CL-ACRWA algorithm and Dijkstra algorithm, the average delay, average delay jitter and average packet loss rate of satellite optical network are reduced. The wavelength utilization is improved. The network congestion probability of high-priority services is reduced.

关 键 词：卫星光网络 波长路由算法 软件定义网络 服务质量 蚁群算法 

分 类 号：TN929.1[电子电信—通信与信息系统]