通信学报

• 学术通信 • 上一篇    下一篇

大规模无线传感器网络(ε,δ)?近似计数算法

朱敬华1,2,管学敏1,2   

  1. 1. 黑龙江大学 计算机科学技术学院,黑龙江 哈尔滨 150001; 2. 黑龙江省数据库与并行计算重点实验室,黑龙江 哈尔滨 150001
  • 出版日期:2013-06-25 发布日期:2013-06-15
  • 基金资助:
    国家自然科学基金青年基金资助项目(61100048);哈尔滨市科技创新人才专项基金资助项目(2011RFQXG028);黑龙江省高校科技创新团队建设计划基金资助项目(2013TD012)

(ε,δ)-approximate counting algorithm for large scale wireless sensor networks

  • Online:2013-06-25 Published:2013-06-15

摘要: 研究了大规模无线传感器网络中的近似计数问题,提出2个基于数字二叉树(DBT, digital binary tree)协议的近似计数算法DBT-ACA和DBT-BACA。算法能够以 的时间复杂性返回 -精度保证的近似计数结果。DBT-BACA采用了二分搜索、逐层转发和延迟响应等技术,有效地减少了查询时间和数据通信量。理论分析和实验结果表明,提出的算法在近似结果的精准度、时间效率和能量开销等方面均优于现有的近似计数算法。

Abstract: The problem of approximate counting for large scale wireless sensor networks was studied. Two approximate counting algorithms, DBT-ACA and DBT-BACA, based on DBT (digital binary tree) protocol were also proposed. The algorithms presented could attain the counting result in time while meeting the accuracy requirement. DBT-BACA exploits binary search, level-by-level forwarding and delay response technique to effectively reduce the query delay and transmission cost. Theoretical analysis and experimental results show that the proposed algorithms outperform existing approaches in terms of estimation accuracy, time efficiency and energy cost.

No Suggested Reading articles found!