计算机工程

• 开发研究与工程应用 • 上一篇    下一篇

一种无线自组网故障检测算法

徐振朋1,沈 浩2,曾玮妮2   

  1. (1. 杰瑞深软科技有限公司,江苏连云港222061; 2. 江苏自动化研究所,江苏连云港222061)
  • 收稿日期:2014-08-28 出版日期:2015-02-15 发布日期:2015-02-13
  • 作者简介:徐振朋(1983 - ),男,高级工程师、博士,主研方向:普适计算,分布式计算;沈 浩,工程师、硕士;曾玮妮,高级工程师、博士。
  • 基金项目:
    国家自然科学基金资助项目(61303045);江苏省自然科学基金资助项目(BK2012237)。

A Failure Detection Algorithm for Ad Hoc Network

XU Zhenpeng 1,SHEN Hao 2,ZENG Weini 2   

  1. (1. JARI Deepsoft Technology Co. ,Ltd. ,Lianyungang 222061,China;2. Jiangsu Automation Research Institute,Lianyungang 222061,China)
  • Received:2014-08-28 Online:2015-02-15 Published:2015-02-13

摘要: 针对无线自组网的拓扑结构,设计一种基于分簇的无线自组网节点故障检测架构和对应的故障检测算法。分簇时分别确定主用簇和备用簇管理节点,冗余簇管理节点负责对内部成员实施异常检测,给出故障检测模块的心跳发送、心跳监控、心跳预判与实时调整机制,通过增加心跳预判实时调整机制,确保算法能够动态适应自组网易变的拓扑结构,并通过备用簇管理节点和簇间共享异常信息机制,提高系统故障检测的可靠性。利用仿真实验对故障检测机制的性能进行评估,结果表明,提出的故障检测算法具备较好的检测准确率,能够有效满足上层应用在系统可靠性设计方面的需求。

关键词: 无线自组网, 容错, 节点故障, 故障检测, 心跳预判

Abstract: A failure detection architecture and algorithm based on clustering are proposed according to the topology of Ad Hoc networks. The active and the backup cluster manager are designated respectively. The exception detection function of the members is implemented by the selected redundancy cluster managers. The sending,monitoring,prediction and updating process of the heartbeat message are designed for fault detection. The updating method of the heartbeat prediction is added to fit the variable topology of Ad Hoc networks dynamically. Through the backup cluster manager and the exception data shared mechanisms among clusters,the system fault detection reliability is improved. The proposal is evaluated by the simulation. As a result,the proposed failure detection mechanism achieves a high accuracy,and is capable of the requirement of the top application design for the system reliability.

Key words: Ad Hoc network, fault tolerance, node failure, fault detection, heartbeat anticipation

中图分类号: