一种新的星座网络多业务类QoS路由算法  

A novel multi-class QoS routing for satellite constellation networks

作　　者：蒋文娟[1] 宗鹏[2] 

机构地区：[1]南京航空航天大学电子信息工程学院,江苏南京210016 [2]南京航空航天大学航天学院,江苏南京210016

出　　处：《江苏大学学报（自然科学版）》2013年第4期428-434,共7页Journal of Jiangsu University：Natural Science Edition

基　　金：国家“863”计划项目(2010AAxxx0404,2009AAxxx404,2008AAxxx405)

摘　　要：针对多媒体业务具有不同QoS要求的问题,提出了一种新的多业务类QoS星座网络路由算法,其目标是在保证高优先级业务性能的同时,提高低优先级业务的性能,从而高效利用网络整体资源.该算法以多种QoS要求和动态链路状态为依据,给出分类的链路代价.引入关键链路的概念,并将链路利用率、剩余带宽、期望负载结合起来定义分类的关键链路代价增量,尽可能减少业务类之间的影响,合理分配网络资源.通过VC和Matlab混合编程建立卫星网络和全球业务仿真环境,并对本文和其他三种路由算法进行仿真试验.结果表明,本文算法不仅保证了高优先级业务的平均路径时延、平均阻塞概率以及平均吞吐率,而且低优先级业务的以上性能具有显著提高,从而有效提高了网络负载均衡性和资源利用率.To solve the different QoS requirements of multimedia traffic, a novel multi-class QoS routing algorithm（MQoSR） for LEO satellite networks was proposed. The ultimate goal of MQoSR was to maxi- mize network resource efficiency by ensuring performance of high-priority traffic classes and to improve the excution of low-priority classes. The classification link cost was presented based on multiple QoS requirments and dynamic link status. The concept of critical link was introduced to reduce the influence between traffic classes and allocate network resources efficiently, and the incremental cost of critical link was defined by link utilization, residual bandwidth and load expectation. Simulation environment of satel- lite network and global traffic was built up by VC and Matlab. MQoSR and three other routings were simulated. The experimental results show that the MQoSR not only guarantees average path delay, blocking probability and throughput of high priority class traffic, but also greatly improves these performance of low priority class traffic. The load balance and the network resource utilization are enhanced effectively.

关 键 词：多业务类 服务质量 链路代价 关键链路 代价增量 

分 类 号：TN91[电子电信—通信与信息系统]