【作者】 李林；

【导师】 何彬彬；

【作者基本信息】 电子科技大学 ， 电子信息（专业学位）， 2023， 硕士

【摘要】 森林火险的及时掌握以及林火的早期探测对于森林火灾的精准防控至关重要。目前森林火灾的防控主要采用地面巡防、瞭望哨驻守、远程视频监控和卫星遥感等方式。尽管取得了一定的效果,但均受制于森林复杂的地理环境及恶劣的气象条件。虽然无线传感器网络技术的发展为森林火灾防控带来了新的契机,但仍然存在节点通信距离短、能耗高及网络拓扑不稳定等问题。因此,亟待研制一种适用于复杂林区环境的无线传感器网络系统,设计林区强遮挡环境下低功耗可靠无线通信组网协议,实现感知信息的远距离、稳定传输。本文提出了混合的介质访问控制协议(Media Access Control protocol,MAC)、改进的分级路由协议以及节点定位算法等一系列低功耗组网协议设计方案,为复杂林区环境无线监测系统的低功耗、稳定可靠监测提供了新的技术思路。论文主要工作如下:(1)对林区环境特征及其对无线通信的影响进行分析,提出了林区环境无线传感器网络的组网需求。通过对多种无线通信技术进行对比,将远距离、低功耗的LoRa技术应用在林区监测上,对现有的MAC协议、路由协议和节点定位算法进行改进,旨在优化网络性能、提高数据传输的可靠性,并适应复杂林区环境中的不确定性和动态性。(2)首先结合基于分配和竞争的MAC协议的优点,提出了一种可根据网络负载情况,自适应调整的混合MAC协议。然后提出一种分级的路由协议,该协议以网关节点为根节点构建一颗树,叶子节点沿着生成树路径传输数据包至根节点,经过树中簇头节点的层层融合处理,大大减少原始数据传输量,降低了能耗。最后,设计了系统中节点的软件和云平台,并对系统的工作流程进行了详细阐述。(3)结合林区环境特征,本文提出了一种新方法,旨在改进传统非测距的距离向量跳数定位算法(Distance Vector Hop Localization Algorithm,DV-Hop),以减少在林区环境中应用该算法时的误差。首先使用多通信半径的思路,修正了DV-Hop算法中的最小跳数计算;随后引入最小二乘准则和加权系数,修正了平均跳距计算。通过仿真分析,改进后的算法在定位精度上面优于传统算法。(4)搭建硬件测试平台,测试多个节点的组网效果以及节点的定位精度,并对系统的生命周期及稳定性进行测试。测试结果表明,所提出的林区无线传感器网络系统方案可实现对复杂林区环境的长时间稳定且可靠的监测。更多还原

【Abstract】 The timely grasp of forest fire risk and the early detection of forest fires are crucial to the accurate prevention and control of forest fires.At present,forest fire prevention and control mainly adopts ground patrol,lookout post stationed,remote video monitoring and satellite remote sensing.Although certain results have been achieved,they are all constrained by the complex geographical environment of the forest and the harsh meteorological conditions.Although the development of wireless sensor network technology has brought new opportunities for forest fire prevention and control,there are still problems such as short communication distance of nodes,high energy consumption and unstable network topology.Therefore,it is urgent to develop a wireless sensor network system applicable to the complex forest environment,design a low-power reliable wireless communication networking protocol under the strong shading environment in forest areas,and realize the long-distance and stable transmission of sensory information.In this thesis,a series of low-power networking protocol design schemes such as hybrid Media Access Control protocol(MAC),improved hierarchical routing protocol and node localization algorithm are proposed to provide a new technical idea for low-power,stable and reliable monitoring of wireless monitoring system in complex forest environment.The main work of the thesis is as follows:(1)The environmental characteristics of forest areas and their impact on wireless communication are analyzed,and the networking requirements of wireless sensor networks in forest environments are proposed.By comparing various wireless communication technologies,the long-range and low-power LoRa technology is applied to forest area monitoring.Improvements are made to existing MAC protocols,routing protocols,and node localization algorithms with the aim of optimizing network performance,enhancing data transmission reliability,and adapting to the uncertainties and dynamics of complex forest environments.(2)Firstly,a hybrid MAC protocol that can be adaptively adjusted according to the network load is proposed by combining the advantages of MAC protocols based on allocation and competition.Then a hierarchical routing protocol is proposed.The protocol constructs a tree with a gateway node as the root node and transmits packets along the spanning tree path to the root node,which greatly reduces the amount of raw data transmission and reduces energy consumption after the layer-wise fusion process of cluster head nodes in the tree.Finally,the software and cloud platform of the nodes in the system are designed,and the workflow of the system is elaborated.(3)Combined with the characteristics of forest environment,this thesis proposes a new method aiming to improve the traditional non-ranging Distance Vector Hop Localization Algorithm(DV-Hop)in order to reduce the error when applying the algorithm in forest environment.Firstly,the minimum hop count calculation in the DV-Hop algorithm is modified using the idea of multiple communication radii;subsequently,the least squares criterion is introduced and the average hop distance calculation is modified.Through simulation analysis,the improved algorithm outperforms the traditional algorithm in terms of positioning accuracy.(4)A hardware test platform is built to test the networking effect of multiple nodes and the positioning accuracy of the nodes,and to test the life cycle of the system.The test results show that the proposed wireless sensor network system solution in forest area can achieve stable and reliable monitoring of complex forest environment for a long time.更多还原