您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(理学版)》

J4 ›› 2011, Vol. 46 ›› Issue (11): 28-32.

• 电子技术与信息 • 上一篇    下一篇

基于蚁群算法的多源组播流量均衡的研究

王另秀,曹叶文*   

  1. 山东大学信息科学与工程学院, 山东 济南 250100
  • 收稿日期:2010-12-31 出版日期:2011-11-20 发布日期:2011-11-30
  • 通讯作者: 曹叶文(1965- ),男,博士,教授,主要研究方向为多载波/多码传输、移动IP组播协议等. Email:ycao@sdu.edu.cn
  • 作者简介:王另秀(1984- ), 女,硕士研究生,主要研究方向为多媒体通信与网络传输技术.Email:wanglingxiu02@yahoo.cn
  • 基金资助:

    国家自然科学基金资助项目(60872023)

A load distribution algorithm based on an ant colony for multi-source multicast networks

WANG Ling-xiu, CAO Ye-wen*   

  1. School of Information Science and Engineering, Shandong University, Jinan 250100,  Shandong, China
  • Received:2010-12-31 Online:2011-11-20 Published:2011-11-30

摘要:

针对现有组播路由技术因路由单一而导致的不能满足多源组播网络中流量均衡的问题,基于蚁群算法提出了一种组播流量均衡的方法——LDA(load distribution algorithm)。LDA主要包括选择候选路由和组播调度两个模块,通过与常用的特定源组播路由协议(PIMSSM)相结合,从整体上考虑均衡网络负载的同时,一方面减小了组播数据包传递的时延,另一方面减小了丢包率。仿真实验结果表明,在PIMSSM的基础上,该方法能有效提高网络资源的利用率,降低组播数据传输时因排队造成的过大的时延和丢包率。

关键词: 组播;蚁群算法;流量均衡

Abstract:

Due to the problem that IP multicast protocols tended to construct a single minimum spanning tree for a multicast source (i.e., group), which can not balance the resource allocation of multicast networks, an ant colony-based load balancing algorithm for multicast networks called the load distribution algorithm (LDA) was proposed. The proposed LDA mainly consisted of two parts: the Selecting Candidate Path and Multicast Scheduling. PIM-SSM (Protocol-Independent Multicast Single-Source Multicast) with the LDA can balance network traffic distribution and meanwhile maintain less packet loss and average delay in the case of co-existing multiple multicast sources. Simulation comparisons between PIM-SSM with the LDA and the original PIMSSM, showed that higher network utilization was achieved in PIM-SSM with the proposed LDA, while maintaining less average end to end delay where there were bottleneck effects.

Key words: multicast; ant colony algorithm; load distribution algorithm (LDA)

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!