摘要
为了克服基于端到端单向时延的拓扑推断算法中需要时钟同步的缺点,根据端到端时延抖动的定义和特点提出了拓扑推断中端到端时延抖动的四元分组列车测量方法和基于端到端时延抖动的拓扑推断算法,其中端到端时延抖动的测量不需要节点间的时钟同步,并且实现简单。分析了基于端到端时延抖动推断网络拓扑的可行性和正确性,通过N S2进行了仿真。仿真结果表明,基于时延抖动推断拓扑结构的效果比基于端到端单向时延推断拓扑结构的效果好。
In order to reduce the limit of time synchronization in network topology inference algorithm based on one way delay, a jitter measurement method named 4-packets train and a topology inference algorithm based on delay jitter were put forward. The jitter measurement method in topology inference was designed based on the definition and characteristics of jitter which did not require time synchronization. The feasibility and correctness of topology inference algorithm based on jitter were analyzed theoretically. Then the algorithm was validated through simulations by NS2. The results prove that the topology inference based on delay jitter can infer network topology better than that one based on one way delay.
出处
《解放军理工大学学报(自然科学版)》
EI
2008年第5期483-487,共5页
Journal of PLA University of Science and Technology(Natural Science Edition)
基金
国家863计划资助项目(2007AA01Z418)
关键词
网络层析成像
时延抖动
拓扑推断
network tomography
delay jitter
topology inference