SIA OpenIR  > 工业信息学研究室
面向工业应用的无线传感器网络时间同步算法研究
Alternative TitleTime Synchronization Algorithms in Wireless Sensor Network for Industrial Application
汪付强1,2
Department工业信息学研究室
Thesis Advisor于海斌 ; 曾鹏
ClassificationTP212
Keyword无线传感器网络 工业应用 时间同步 高精度 鲁棒性 节能
Call NumberTP212/W27/2011
Pages122页
Degree Discipline机械电子工程
Degree Name博士
2011-05-26
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract工业无线传感器网络技术改变了现有系统控制信息传递的方式,是继现场总线之后,工业控制领域的又一个热点技术。基于无线技术的信息传递方式,使得测控系统具有低成本、易使用、易维护等优点,在工业控制领域具有广阔的应用前景。 时间同步是工业无线传感器网络研究的重要内容之一。当前时间同步技术的研究主要为了满足大规模、自组织、动态网络的组网需要,针对无线传感器网络稳定性、收敛性、鲁棒性、能量感知、可扩展性等特点进行协议的设计和分析。工业无线传感器网络的作为一种特殊的无线传感器网络对时间同步的研究提出了新的要求。首先,工业无线传感器网络有较高的实时性,必然要求更高的同步精度和稳定的时间同步;其次,工业现场存在大量的电磁噪声,其他无线设备的射频干扰,以及特殊环境下高温高湿和遮挡产生的多径干扰等影响,导致的丢包问题相对传统无线传感器网络更加严重,这就要求时间同步具有高可靠性;再者,工业现场设备都要求数月甚至数年的持续工作,必然对无线节点的功耗提出了更严格的要求,因此需要时间同步能够以更低的功耗满足用户需求。工业无线传感器网络的特殊性使得现有的时间同步协议不能直接用于工业无线传感器网络中。论文针对工业无线监测应用的需求和特点展开时间同步的研究工作,主要包括以下几方面内容。 论述了工业无线传感器网络时间同步技术设计所面临的挑战性问题,系统地总结了已有研究成果与不足,具体分析了在工业过程监测环境下无线传感器网络的特点和时间同步需要重点解决的问题。 针对工业现场特殊环境提出了基于层次结构的鲁棒时间同步方法。通过基于跳数的层次结构避免了传统时间同步构造网络拓扑结构的过程,随机时间源方法使得在干扰较多的工业环境下降低了节点间同步失效几率,而重同步技术能够快速检测和恢复节点的同步失效。 提出了基于簇结构的主动和被动结合双向同步方法,解决了传统双向同步报文交互频繁的问题,通过最大似然估计校正节点间的相位偏差和频率偏差,使得同步算法获得单向同步报文代价下较高的同步精度。 提出了基于预测补偿的时间同步周期优化方法,通过同步误差预测补偿算法降低节点间的累积误差,并根据用户精度需求动态调整节点间同步周期,从而达到在设定精度下的最大化同步周期,在特定需求下最大化同步周期,大大降低了同步开销。 介绍了工业无线测控系统,并根据系统需求建立了工业无线传感器网络时间同步体系。针对传统的同步精度测试手段受网络通信资源限制,本文设计了基于同步误差映射的时间同步精度测试平台,通过FPGA查询端口状态的方式获得实时的同步误差,避免了同步精度测试对正常网络的影响。针对工业现场测试的局限性,通过对典型工业现场无线通信环境的实地测量,建立了工业无线环境模拟实验平台,便于对工业无线传感器网络的快速开发和测试。最后在本文时间同步算法体系的基础上实现了基于无线Hart协议的工业无线测试系统,全面测试了系统性能并进一步验证了本文算法的有效性。
Other AbstractIndustrial wireless sensor network technology (IWSN) which has changed the information transmission way for existing control system is another hotspot technology in industrial control field following the fieldbus. Based on wireless technology information transfer mode, the measurement and control system, which has advantages of low cost, easy to use and easy maintenance, has wide application prospects in the industrial control area. Time synchronization is one of the important contents in industrial wireless sensor network research. The current time synchronization technology research is mainly in order to meet the large-scale, self-organizing, and dynamic network networking need. The design and analysis of the research has the characteristics of stability, convergence, robustness, energy perception and scalability in wireless sensor network. Industrial wireless sensor network, as a special kind of wireless sensor network (WSN), puts forward new requirements for the study of time synchronization. First, industrial wireless sensor network has the performance of high real-time, which leads to higher accuracy and stability of time synchronization. Second, industrial environment sensors may be subject to electromagnetic noise,RF interference from other wireless equipments and even high temperature and humidity, multi-path interference caused by occlusion in special surroundings, which makes the packet loss problem more serious than traditional wireless sensor network. So the time synchronization should have high reliability. Moreover, industrial field devices require months or years continue to work, and it is inevitably bring necessary for wireless nodes with more stringent requirements of the power consumption, so time synchronization should work at lower power consumption to meet user needs. The particularity of industrial wireless sensor network makes wireless sensor network time synchronization protocol can't be directly used in industrial wireless sensor network. This paper does research on time synchronization according to industrial wireless monitoring application requirements and characteristics.The main contributions are summarized as follows. The research on WSN synchronization is surveyed, while the challenges and design issues that affect synchronization process in industrial process monitoring applications are discussed. According to the character of industrial wireless networks, we design a robust time synchronization protocol based on level structure in this paper. The level structure based on network hops can avoid topology construction in traditional synchronization methods. We also employ random time source to decrease the probability of synchronization failure in industrial environment with much interference and re-synchronization mechanism to provide rapid detection and recovery. The combination of active and inactive two-way synchronization based on clustering architecture is proposed in this paper. The clustering architecture pattern solves the frequent packet interaction for the traditional two way methods. The phase offset and frequency offset are corrected through the maximum likelihood estimation. Based on these considerations, we can acquire high accuracy and one way synchronization message cost. In this paper, we proposed a synchronization cycle optimization protocol based on error prediction and compensation algorithm which are designed to decrease the accumulative error between nodes. By this method, the synchronization cycle of wireless networks can be adjusted dynamically as corresponding to fixed time error limit. The increase of synchronization cycle is equal to energy-saving. Industrial wireless measurement and control system is introduced and time synchronization system is established for industrial wireless sensor network in this paper. We design a synchronization precision testing platform based on synchronization error mapping. We get real time synchronization error from inquiring the port status way through FPGA, to avoid the influence to normal network when testing the synchronous precision. For the inconvenience of field test we establish a simulation experiment platform to simulate industrial wireless environment that accelerate the development and testing of industrial wireless sensor network. At last we establish a testing system based on the Wireless Hart Protocol and our time synchronization system. For a long time roundly test we collect the data to make further performance evaluation of the network and our synchronization mechanism in industrial environment.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/9361
Collection工业信息学研究室
Affiliation1.中国科学院沈阳自动化研究所
2.中国科学院研究生院
Recommended Citation
GB/T 7714
汪付强. 面向工业应用的无线传感器网络时间同步算法研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2011.
Files in This Item:
File Name/Size DocType Version Access License
面向工业应用的无线传感器网络时间同步算法(2545KB) 开放获取LicenseApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[汪付强]'s Articles
Baidu academic
Similar articles in Baidu academic
[汪付强]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[汪付强]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.