面向光缆制造的无线传感器网络时间同步研究

发布时间：2019-03-19 15:21

【摘要】：随着物联网技术飞速的发展,使工业生产越来越智能化、信息化。在传统光缆制造车间内需要大量操作人员,这样需要消耗大量人力、财力等成本。通过在光缆制造车间内布置传感器节点使得数据采集越来越方便、高效。光缆制造监测系统中数据采集、融合、处理等都是建立时间同步的基础上,因此时间同步是一项重要的技术支持。由于光缆制造是在室内进行的,成熟的GPS无法精确的获取全球标准时间UTC。将经典时间同步算法应用在室内对传感器网络时间进行同步虽然可行,但经典算法随网络节点增多会增加网络负担、误差积累等问题不适应用于环境复杂的光缆制造网络,因此本文提出移动式线性频率互异脉冲耦合振荡器算法。论文主要研究内容如下:介绍光缆制造环境下使用无线传感器网络现状、关键技术支持,进而引出关键的时间同步技术以及时间同步问题。分析了无线传感网络时间同步原理和一些经典的时间同步算法,由于经典时间同步算法存在不足,则提出脉冲耦合振荡器算法(Pulse coupled oscillator algorithm,PCOs)。通过对PCOs算法进行分析研究,因PCOs算法是非线性且会增加网络负担,从而提出线性PCOs算法。由于在无线传感器网络中各传感器晶振频率不相同且随时间推移晶振温度升高造成漂移现象,利用最小二乘法曲线拟合(least-squares curve fitting,LSCF)能有效解决晶振频率问题提高同步时间精度。为全方位监测网络环境,提出移动式线性频率互异PCOs算法。通过Matlab软件仿真验证节点最后能较好地达到同步,同时比较了网络参数速度v、耦合强度ε、节点n在相对区间内与同步时间收敛速度成正比。在硬件系统中主要采用嵌入式芯片STM32F051作为控制中心以及CC2530开发套件作为传感节点硬件;在软件系统中应用IAR对程序进行编程设计。通过验证,本文设计的算法能使网络节点之间的时间同步误差值在[0,10]us区间内,表明网络节点时间能有效达到同步。从而可以得出将移动式线性频率互异PCOs算法应用到网络中更能有效解决光缆制造中无线传感器网络时间同步问题。
[Abstract]:With the rapid development of Internet of things technology, industrial production is becoming more and more intelligent and information-based. A large number of operators are required in the traditional optical cable manufacturing workshop, which requires a large amount of manpower, financial resources and other costs. It is more and more convenient and efficient to collect data by arranging sensor nodes in the optical fiber optic cable manufacturing workshop. Data acquisition, fusion and processing in optical cable manufacturing monitoring system are all based on the establishment of time synchronization. Therefore, time synchronization is an important technical support. Because optical cable manufacturing is carried out indoors, mature GPS is unable to accurately obtain the global standard time UTC. It is feasible to use the classical time synchronization algorithm to synchronize the time of sensor network indoors, but the classical algorithm will increase the network burden with the increase of network nodes, and the problem of error accumulation is not suitable for the optical fiber optic cable manufacturing network with complex environment. Therefore, a mobile linear frequency reciprocal pulse coupled oscillator algorithm is proposed in this paper. The main contents of this paper are as follows: the status quo and key technical support of wireless sensor networks in optical cable manufacturing environment are introduced, and then the key technology of time synchronization and time synchronization are introduced. The time synchronization principle of wireless sensor network and some classical time synchronization algorithms are analyzed. A pulse coupled oscillator (Pulse coupled oscillator algorithm,PCOs) algorithm is proposed because of the shortcomings of the classical time synchronization algorithm. By analyzing and studying the PCOs algorithm, the linear PCOs algorithm is proposed because the PCOs algorithm is nonlinear and will increase the network burden. Because the crystal oscillator frequency of each sensor is different in wireless sensor networks and the crystal oscillator temperature increases over time, the least squares curve fitting (least-squares curve fitting,) is used to simulate the drift phenomenon. LSCF) can effectively solve the problem of crystal oscillator frequency and improve the precision of synchronization time. In order to monitor the network environment in an all-round way, a mobile linear frequency reciprocal PCOs algorithm is proposed. The simulation of Matlab software shows that the node can achieve synchronization well at last. At the same time, the speed of network parameter v, coupling intensity 蔚 and node n are proportional to the convergence rate of synchronization time in the relative interval. In the hardware system, the embedded chip STM32F051 is used as the control center and the CC2530 development kit is used as the sensor node hardware, and the program is programmed with IAR in the software system. Through verification, the algorithm designed in this paper can make the time synchronization error between nodes in [0, 10] us, which shows that the time synchronization of network nodes can be achieved effectively. Therefore, it can be concluded that the time synchronization problem of wireless sensor networks in optical cable manufacturing can be solved more effectively by applying the mobile linear frequency reciprocal PCOs algorithm to the network.
【学位授予单位】：安徽理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【相似文献】