水下传感器网络时间同步技术研究

发布时间：2018-07-11 15:47

本文选题：水下传感器网络 + 时间同步　；参考：《中国海洋大学》2014年硕士论文

【摘要】：随着信息技术和海洋科技的飞跃发展，作为一种新兴的海洋信息采集技术，水下声学传感器网络（Underwater Acoustic Sensor Networks, UASN）越来越受到重视，被广泛应用于海洋环境监测、海洋污染监控、海洋资源勘测、水下目标的探测、跟踪和定位等领域。该网络的主体由水面基站和水下传感器节点组成，节点部署范围大、密度低；节点之间采用水下声波通信，而水声通信的链路质量受海洋环境变化的复杂因素影响，具有低传播速度、高延迟、延迟时变、高误码率、低带宽、多途效应严重及高能耗等特点。时间同步是传感器网络中一项重要的支撑技术，它为协调跨多节点的传感与驱动提供统一的参考时钟，是节点协同工作的基础，传感器网络的数据融合、时分复用调度、节点定位和协同休眠等关键技术都依赖于节点间的时间同步。由于水下传感器网络采用声波通信，，传播速度低且通信链路较长，并考虑到节点在水下随海流移动，导致了巨大的传播延迟和时空不确定性，使得传统的陆上无线传感器网络时间同步算法不能直接应用于水下。因此，UASN时间同步中必须同时考虑延迟时变性和节点相对运动造成的影响，这为同步算法研究带来了新的挑战。目前，国内外对于UASN时间同步算法的研究，主要采用基于单信标单跳的点到点同步机制，即假定整个网络或网络的一个分簇中，每个节点都存在到信标节点的单跳路由，通过与信标节点的多次消息交互和计算，即可完成全网时间同步。但是，由于海洋监测任务涉及范围较大，实际中水下节点多采用0.5km~2km的长间距部署，节点间通信方式一般采用基于多跳传输的机会路由。所以，现有的点到点同步算法，难以适应水下传感器网络的实际部署要求。本文为解决多跳网络中同步精度随跳数降低的问题，研究和设计基于多信标节点和多跳传输的多源同步消息融合算法，充分利用多源同步消息的时间有效性，以提高整个网络的同步精度。本文提出水下传感器网络多源时间同步算法MSTS（Multi-Source Time Synchronization），采用仿真实验探究方法评估路径跳数对于同步消息时间有效性的影响，在多次消息交互中融合不同有效性的同步消息，运用以消息时间置信度为权的加权线性拟合，估计节点时钟频率偏斜和相位偏差。同时，为准确评估水下同步算法的性能，本文研究和开发了水下传感器网络仿真系统（Underwater Sensor Network Simulator，UWSN-Sim），建立水声通信模拟环境及节点运动模型，模拟多时间基准节点条件下的多跳同步消息传输。利用UWSN-Sim仿真系统，设置不同的仿真场景和通信参数，通过大量仿真实验模拟水下传感器网络不同的拓扑结构和部署规模，验证了MSTS算法在多跳拓扑下具有较高的同步精度；并且，在大规模部署的多跳网络中，与逐跳迭代同步的MU-Sync算法相比较，MSTS算法以较低的平均消息开销和更短的总体运行耗时，达到了更高数量级的同步精度。
[Abstract]:With the rapid development of information technology and marine science and technology, as a new marine information acquisition technology, Underwater Acoustic Sensor Networks (UASN) has been paid more and more attention. It is widely used in marine environment monitoring, marine pollution monitoring, marine resources survey, underwater target detection, tracking and positioning. The main body of the network is composed of water surface base station and underwater sensor node, with large deployment range, low density, underwater acoustic communication between nodes, and the link quality of underwater acoustic communication is influenced by complex factors of ocean environment change, with low propagation speed, high delay, delay time change, high error rate, low bandwidth and multipath effect. The characteristics of severe and high energy consumption.
Time synchronization is an important support technology in sensor networks. It provides a unified reference clock for coordinating the sensing and driving of multiple nodes. It is the basis of the joint work of nodes. The key technologies of sensor networks, such as data fusion, time division multiplexing, node location and cooperative dormancy, depend on time synchronization between nodes. Underwater sensor networks adopt sound wave communication with low propagation speed and long communication links. Considering that nodes move with sea currents under water, it leads to huge propagation delay and time and space uncertainty, which makes the traditional time synchronization algorithm of terrestrial wireless sensor network cannot be directly used in water. Therefore, the time synchronization of UASN must be simultaneous. Considering the effect of delay time variation and node relative motion, this brings new challenges to synchronization algorithm research.
At present, at home and abroad, the research of UASN time synchronization algorithm mainly adopts the point to point synchronization mechanism based on single beacon single hop, which assumes that each node has a single hop route to the beacon node in a cluster of network or network, and can complete the whole network time synchronization through multiple message interaction and computation with the beacon node. Because the ocean monitoring task involves a large range, the actual underwater nodes are mostly deployed in long distance between 0.5km~2km, and the communication mode between nodes generally adopts the opportunity routing based on multi hop transmission. Therefore, the existing point to point synchronization algorithm is difficult to adapt to the actual deployment requirements of underwater sensor networks.
In order to solve the problem of reducing the synchronization accuracy with the number of hops in multi hop networks, this paper studies and designs a multi source synchronization message fusion algorithm based on multi beacon and multi hop transmission. It makes full use of the time validity of multi source synchronization messages to improve the synchronization accuracy of the whole network. This paper proposes a multi source time synchronization algorithm MSTS for the underwater sensor network. (Multi-Source Time Synchronization), the simulation experiment is used to evaluate the effect of path hops on synchronization message time validity. The synchronization messages with different effectiveness are fused in multiple message interaction, and the weighted linear fitting with the weight of message time confidence is used to estimate the node clock frequency skew and phase deviation.
At the same time, in order to accurately evaluate the performance of underwater synchronization algorithm, this paper studies and develops an underwater sensor network simulation system (Underwater Sensor Network Simulator, UWSN-Sim), establishes a simulation environment of underwater acoustic communication and a node motion model, and simulate multi hop synchronous message transmission under the condition of multi time datum nodes. A UWSN-Sim simulation system is used. Different simulation scenes and communication parameters are set up, and the different topology and deployment size of underwater sensor network are simulated by a large number of simulation experiments. It is verified that the MSTS algorithm has high synchronization accuracy under multi hop topology. And, in the multi hop network of large-scale deployment, the MSTS algorithm is compared with the MU-Sync algorithm which is synchronized with the hop by hop iteration. With lower average message overhead and shorter overall operation time, the synchronization accuracy is higher.
【学位授予单位】：中国海洋大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.3;TP212.9

【参考文献】