电信科学

• • 上一篇    下一篇

移动Ad Hoc网络环境下基于风险的 高效可靠的服务组合方法研究

胡海洋,吕 倩,俞东进,赵格华   

  1. 杭州电子科技大学计算机学院 南京大学软件新技术国家重点实验室;杭州电子科技大学计算机学院;杭州电子科技大学计算机学院;香港中文大学计算机科学与工程系
  • 出版日期:2012-11-15 发布日期:2012-11-15
  • 基金资助:
    国家自然科学基金资助项目(No.60903053,No.61272188),浙江省重大科技专项基金资助项目(No.2008C11099-1),浙江省自然科学基金资助项目(No.Z1100822,No.LY12F02005),南京大学软件新技术国家重点实验室开放基金资助项目(No.KFKT2011B07)

Research on Efficient and Reliable Service Composition in Mobile Ad Hoc Networking Based on Risk Analysis

Hu Haiyang,Lv Qian,Yu Dongjin and Zhao Gehua   

  1. School of Computer Science and Technology, Hangzhou Dianzi University State Key Laboratory for Novel Software Technology of Nanjing University;School of Computer Science and Technology, Hangzhou Dianzi University;School of Computer Science and Technology, Hangzhou Dianzi University;Department of Computer Science and Engineering, Chinese University of Hong Kong
  • Online:2012-11-15 Published:2012-11-15

摘要: 在移动Ad Hoc网中,由于节点以及网络拓扑结构的动态性,服务组合性能往往比在静态网络环境中更不稳定。为了提高服务组合的可靠性,大量的研究工作关注最小化服务组合的中断次数以及服务组合的重组,而对于在服务进行组合之前就预先判断服务的可靠性的研究较少。本文旨在利用移动预测技术结果,在服务组合使用之前寻找具有较低失效风险的服务组合方案,提出了失效风险模型以衡量服务组合的可靠性,并且在该模型的基础上设计了3种算法用于求解服务组合。本文还进行了相应的实验来评估所提出算法的特点。

Abstract: In the mobile ad hoc networks, the reliability of service composition is always less than the one in static network environment because of the mobility of nodes and the dynamics of network topology. In order to improve the reliability of service composition, current research works mainly focus on minimizing times of interruption and the services re-composition, and pay little attention to estimating the reliability before composition. By taking advantages of the mobility prediction for mobile nodes, this paper proposed approaches to find the service composition scheme with low failure risks before the composition is performed. It proposed a failure risk model to estimate the reliability of service composition, and then presented three algorithms for performing service compositions based on the reliability estimated. Finally, by comparing with the optimal scheme, we conducted experiments to evaluate the features of our methods.

No Suggested Reading articles found!