SIA OpenIR  > 工业信息学研究室
Alternative TitleCapacity Analysis and Transmission Scheduling for Industrial Wireless Sensor Networks
Thesis Advisor于海斌 ; 梁炜
Keyword工业无线传感器网络 容限分析 传输调度 实时性 可靠性
Call NumberTP212/Z35/2011
Degree Discipline机械电子工程
Degree Name博士
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract系统归纳了无线传感器网络中传输调度问题的评价指标、分类方法、研究现状,并指出了存在的问题和未来研究重点。 基于IEEE 802.15.4信标调度机制中的典型算法——超帧持续期信标调度算法,充分利用工业应用周期性数据传输的特点,基于概率论方法分析了树簇型IEEE 802.15.4网络的容量;并进一步分析了影响树簇拓扑网络容量的因素,得到了网络规模的增长速度远远大于网络容量增长速度的结论。仿真实验验证了理论推导的正确性。该研究解决了传统网络容量分析方法脱离具体传输协议、忽略传输半径对网络性能影响而带来的结果理想化问题,为实际的网络性能分析和协议设计提供了理论上的支持。 针对工业无线传感器网络的实时通信要求,分别研究了线簇型网络和树簇型网络完成汇聚传输所需时隙数和信道数的下限值;并分析了网络结构、网络规模和数据更新周期对于时隙数和信道数下限值的影响。在此基础上,基于理论分析的下限值,提出了面向线簇型网络和树簇型网络的实时汇聚传输调度算法。仿真和物理实验验证了所提算法在实时性和低开销等方面的优势。 针对现场环境复杂多变的特性和工业应用的可靠性要求,提出一种基于低开销信道状态认知的自适应可靠传输调度算法。首先提出了基于动态规划方法的在线信道认知和建模,优化了模型准确度和信道探测代价;在此基础上,提出了基于全局时隙调度和局部信道调度的可靠传输调度算法。仿真和实验结果表明,该算法一方面在提供环境适应性的同时克服了现有在线信道建模方法忽略探测代价的弊端;另一方面在分担全局传输调度负担的同时尽可能保证了调度的最优性,提高了资源利用率,保证了报文传输的可靠性。 针对传输调度受功率控制和路由影响的问题,提出一种具有环境适应性的多信道路由、传输调度和功率控制联合多目标优化的算法。首先基于SINR物理干扰模型,设计了考虑截止期约束的功率控制算法,解决了传输半径和功率固定带来的高能耗和强干扰问题。其次,提出了报文重传次数的优化方法,在保证可靠性的同时提高了资源利用率。基于优化的传输半径、功率、数据率以及重传次数,提出了一种路由和传输调度联合优化算法,解决了网络拓扑和网络载荷动态变化导致的高控制开销的问题。最后,利用李雅普诺夫稳定性直接判据证明了算法的收敛性,深入分析了路由算法、信道数、网络规模、数据率和功率对于网络性能的影响,并进行了仿真验证。 总之,论文对工业无线传感器网络的容限分析和传输调度方法进行了较为深入的研究和探讨,旨在为无线传感器网络的实际应用,特别是工业应用,提供强有力的理论依据和技术支持。
Other AbstractAfter the wide use of Internet, Wireless Sensor Networks (WSNs) have become another hot technology, which have significant impact on the style of human life in the twenty-first century. Industry is one of the important applications for WSNs. Industrial Wireless Sensor Networks (IWSNs) have many advantages, such as low-cost, easy-use, and easy-maintenance. Therefore, IWSNs become a revolutionary technology; it could reduce the cost of industrial monitoring & control systems and extend the application scopes. Compared with other applications of WSNs, IWSNs have the requirements of real-time requirements, such as real-time and reliability. It is a challenge to meet these requirements in the harsh industrial environment. Transmission scheduling is a key approach to meet industrial application requirements, because it can solves the problem how to effectively utilize wireless communication resources reasonably and effectively. The transmission scheduling for IWSNs is a multiple-object optimizing problem and also an NP-complete problem. In addition, there are complex spatial-time constraints because of multi-hop communication tasks, hard timeliness requirement, and location-dependent channel resources; there is combination explosion problem because of large network scale and periodic tasks; centralized approaches can not satisfy the application requirements due to large-scale distributed system, while there are deterministic and global optimization issue if use distributed approach; It bring higher challenges for adaptive transmission scheduling because of some uncertain features, such as aperiodic tasks, dynamic channel conditions, and the change of topology and route. All these mentioned factors above limit the use of existing methods and increase difficulties of analysis and solutions. The dissertation studies the capacity and transmission scheduling of IWSNs. The detailed content and the innovative achievements are described below. The concept, characteristics, and technical requirements of WSNs, especially for the industrial WSNs, are represented in the dissertation. Furthermore, the performance metrics, classification, and related research work of the transmission scheduling are summarized and discussed, the future research emphases are pointed out. For a WSN with cluster-tree network topology, we focus on the network capacity of a typical IEEE 802.15.4 beacon scheduling algorithm which is called Superframe Duration Scheduling (SDS). The periodic data transmission is carefully considered and the probability theory is used. We analyze the factors that influence the network capacity; the conclusion shows that the increase speed of network scale is much faster than that of the network capacity. The validation of theoretical derivation is verified by the simulation. Traditional methods do not consider the low-layer transmission protocol and ignore the influence by the transmission radius to the network performance, so their analysis results are too ideal. The research resolves these problems and provides theoretical support for network performance analysis and protocol design. For the real-time requirement of IWSNs, we establish the lower bounds of timeslots and channels when cluster-line routing and the cluster-tree routing finish convergecast communication. The network structure, network scale and data update rates are analyzed for the influences to the lower bounds. Based on the theoretical analysis, we propose the real-time convergecast scheduling algorithms for cluster-line networks and cluster-tree networks. Finally, the advantages of this algorithm are validated by simulation and experiments in the aspects of real-time performance and low complexity. An adaptive and reliable transmission scheduling algorithm is proposed based on low-cost cognition of channel state to meet the requirement of reliable transmission in the industrial complex environment. Firstly, channel modeling is established by using the dynamic programming methods, this can increase the model accuracy and optimize the channel probing cost. Based on this improved channel model, we proposed a hybrid method of global timeslot scheduling and local channel scheduling. Simulation and experiment results show that this algorithm could 1) show adaptability to the dynamic network environment and solve the problem of channel modeling without considering the probing cost; 2) alleviate the burden of global scheduling, guarantee the global optimality, increase the resource utilization, and improve the reliability. The transmission scheduling is influenced by the strategies of power control and routing. In order to solve these problems, we propose a jointly optimizing algorithm for routing, scheduling, and power control, which aims to realize multiple objects and be adaptive to the environment. Firstly, we design a deadline-constrained power control sub-algorithm based on the SINR physical interference model, it can eliminate the high energy-consumption and strong interference caused by constant transmission radius and rate. Secondly, we establish a link-level reliability model based on the expectation theory, based on this model, an optimized algorithm of retransmissions is proposed. This algorithm can guarantee the reliability and increases the utilization of communication resource. Then we jointly optimize the routing and transmission scheduling based on the optimized transmission radius, power, rate, and retransmissions, this transmission scheduling reduces the control overhead which is caused by the dynamic network topology and data stream. The validation is proved theoretically by utilizing the Lyapunove direct stability criteria. Finally, the influence on the network performance by the reliability model, routing algorithm, number of channels, network scale, transmission rate, and power is deeply analyzed. Simulation results indicate the effectiveness of our algorithm in multiple applications. In summary, the dissertation focuses on the research of capacity analysis and transmission scheduling of IWSNs, the result will provide strong theoretical base and technical support for the industrial applications of IWSNs.
Contribution Rank1
Document Type学位论文
Recommended Citation
GB/T 7714
张晓玲. 工业无线传感器网络的容限分析与传输调度方法研究[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
工业无线传感器网络的容限分析与传输调度方(2617KB) 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[张晓玲]'s Articles
Baidu academic
Similar articles in Baidu academic
[张晓玲]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[张晓玲]'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.