English

应用科学学报 ›› 2017, Vol. 35 ›› Issue (6): 717-725.doi: 10.3969/j.issn.0255-8297.2017.06.005

• 通信工程 • 上一篇    下一篇

非平稳与非完美通信条件下的CI分布式算法

匡红梅1, 李伟2   

  1. 1. 首钢工学院 信息工程系, 北京 100041;
    2. 北方工业大学 计算机学院, 北京 100144
  • 收稿日期:2016-09-26 修回日期:2017-03-27 出版日期:2017-11-30 发布日期:2017-11-30
  • 作者简介:匡红梅,讲师,研究方向:信号与信息处理,E-mail:hongmei_kuang@163.com

CI Distributed Algorithm under Non-stationary and Imperfect Communication Conditions

KUANG Hong-mei1, LI Wei2   

  1. 1. Department of Information Engineering, Shougang Institute of Technology, Beijing 100041, China;
    2. School of Computing, North China University of Technology, Beijing 100144, China
  • Received:2016-09-26 Revised:2017-03-27 Online:2017-11-30 Published:2017-11-30

PDF

248

摘要：

在无线传感网中，传感节点需从物理世界中估计一些兴趣参数，因此分布式估计是一个重要问题.为解决这一问题，该文发展了CI（consensus-plus-innovations）分布式算法，并从理论上分析了非平稳条件及非完美通信条件下CI分布式算法的均值收敛性能，说明CI分布式算法是均值收敛的，此外还分析了算法的渐进正态性.研究表明，非平稳条件和非完美通信条件并不影响CI算法的收敛性和渐进正态性，但会影响渐进方差.通过仿真验证了CI分布式算法的合理性.

关键词: 非平稳, 均值收敛, CI算法, 分布式估计

Abstract:

In wireless sensor networks, distributed estimation is an important issue, in which sensor nodes estimate parameters of interest from the physical world. This paper develops a consensus-plus-innovations (CI) distributed algorithm to deal with distributed estimation. The focus is on the mean convergence performance of a CI distributed algorithm under non-stationary and imperfect communication conditions. Theoretical analysis shows that the algorithm is mean convergent and has an asymptotic normality property. Also, non-stationarity and imperfect communication conditions have no effect on the mean convergence performance and asymptotic normality. However, these conditions have an impact on the asymptotic variance of the algorithm. Validity of the CI distributed algorithm is shown by simulation.

Key words: CI algorithm, distributed estimation, non-stationary, mean convergence

中图分类号: