摘要
为克服现有卫星网路由算法的不足,结合单层卫星网负载分布不均匀的特点,提出了一种新型的适用于单层卫星网的基于移动Agent的动态路由算法(SDRA-MA)。该算法通过移动Agent在卫星节点间迁移,收集星际链路时延、卫星纬度等信息。当条件满足时,移动Agent往回迁移,并在每个中间卫星节点基于卫星地理位置计算所探测路径的代价、更新路由表。通过在类Iridium星座中的仿真结果表明,SDRA-MA能够适应网络拥塞,实现负载平衡。尤其在高负载情况下端到端时延、丢包率等指标均优于传统的卫星路由算法。通过复杂度分析得知,SDRA-MA具有较低的通信、存储和计算开销。
To overcome the deficiencies of the existing satellite routing algorithms,satellite network dynamic routing algorithm based on mobile agent（SDRA-MA） was proposed considering the unfair distribution of network traffic.For SDRA-MA,mobiled agents migrated among satellites to gather the inter-satellite link（ISL） cost and the latitude of visited satellite.When conditions permitted,mobile agents migrated back.On each intermediate satellite,mobile agents evaluated the path cost considering the satellite geographical position as well as the ISL cost,and finally updated the routing tables.Simulations on the Iridium-like system show that the proposed algorithm has adaptive abilities to deal with congestion and to achieve load balancing,and that in case of high traffic load,SDRA-MA can achieve better end-to-end delay guarantee and lower packet loss ratio compared to the traditional satellite routing algorithms.Furthermore,results of the complexity analysis demonstrate that SDRA-MA has low requirements on onboard signaling,storage and computation.
出处
《解放军理工大学学报（自然科学版）》
EI
北大核心
2010年第3期255-260,共6页
Journal of PLA University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(60973139
60773041)
国家863计划资助项目(2006AA01Z201
2007AA01Z404
2007AA01Z478)
江苏省自然科学基金资助项目(BK2008451)
现代通信国家重点实验室基金资助项目(9140C1105040805)
江苏省博士后基金资助项目(0801019C)
江苏高校科技创新计划资助项目(CX08B-085Z
CX08B-086Z)
江苏省六大高峰人才项目
