【作者】 李云鹏；

【导师】 刘丽钧；

【作者基本信息】 沈阳工业大学 ， 检测技术与自动化装置， 2021， 硕士

【摘要】 无线传感器网络技术是一门集检测、通信、控制等多种技术于一身的融合技术,作为无线传感器网络技术核心之一的节点定位技术,其在瓦斯监测、森林火灾监测、室内定位、农业物联网等领域有着广泛的应用。节点定位技术是无线传感器网络环境监测的前提,只有知道节点的具体位置,监测到的环境数据才会产生实际价值,但在实际应用中,定位精度与定位覆盖率会受到节点分布不均与环境中电磁干扰等影响,最终影响传感器的监测。因此,对无线传感器网络定位的研究是有着重要现实意义的。近似三角形内点测试(APIT)算法是一种非基于测距的定位算法,它在拥有计算量较小、功耗较低、对硬件依赖程度较弱、定位成本较低等优势的基础上,仍然有着相对较高的定位精度,因此得到了国内外专家学者的广泛关注。本文对非基于测距的APIT定位算法进行了深入地研究,并在APIT定位算法的基础上提出了基于APIT的混合定位算法。本文的主要工作内容如下:(1)对无线传感器网络技术的发展进行了梳理,阐述了无线传感器网络定位技术的研究意义,对国内外关于无线传感器网络定位技术研究做出的工作进行了介绍,了解了无线传感器网络定位技术的研究背景。(2)介绍了Zig Bee无线传感技术、网络拓扑结构及Z-Stack协议栈,讲述了无线传感器网络定位技术及几种常用定位算法的原理,并给出了定位算法的性能指标,通过将几种定位算法进行对比,分析了节点的分布对定位结果的影响。(3)分析了APIT定位算法的不足,并总结了造成不足的原因,然后在APIT定位算法的基础上提出了基于APIT的混合定位算法,混合算法通过三角形比较分割对APIT定位算法中网格扫描阶段进行优化,并结合遗传算法对通讯范围内参考节点数仅有2个的盲节点进行定位,从而提高定位精度与覆盖率。(4)对无线传感器网络定位系统进行了设计,并通过CC2530芯片与IAR Embedded Workbench软件开发环境将其实现。在CC2530中移植Z-Stack协议栈,利用协调器、参考节点以及盲节点组成拓扑网络,将采集到的数据带入算法实现,并通过MATLAB的GUI图形界面展示盲节点位置。为了验证混合算法的有效性,在相同的实验环境下,对其进行了软件仿真与硬件实验,结果显示,硬件实验与软件仿真结果在整体上保持一致的同时,误差也都控制在了合理范围之内,这表明了混合算法不仅具有理论上的可行性,也具有一定的实际应用价值。更多还原

【Abstract】 Wireless sensor network technology is a fusion technology of detection,communication,control and other technologies.As one of the core technologies of wireless sensor network,node localization technology has a wide range of applications in gas monitoring,forest fire monitoring,indoor positioning,agricultural Internet of things and other fields.Node localization technology is the premise of wireless sensor network environment monitoring.Only knowing the specific location of the node,the monitored environmental data will produce practical value.But in practical application,the localization accuracy and localization coverage will be affected by the uneven distribution of nodes and electromagnetic interference in the environment,which will ultimately affect the sensor monitoring.Therefore,the research of wireless sensor network localization is of great practical significance.Approximate triangle interior point test(APIT)algorithm is a range-free localization algorithm.It has the advantages of less computation,lower power consumption,less dependence on hardware and lower localization cost,and still has relatively high localization accuracy.Therefore,it has been widely concerned by experts and scholars at home and abroad.In this thesis,the APIT localization algorithm based on range-free is deeply studied,and a hybrid localization algorithm based on APIT is proposed.The main contents of this thesis are as follows:(1)This thesis combs the development of wireless sensor network technology,expounds the research significance of wireless sensor network localization technology,introduces the research work on wireless sensor network localization technology at home and abroad,and understands the research background of wireless sensor network localization technology.(2)This thesis introduces Zig Bee wireless sensor technology,network topology and Z-stack protocol stack,describes the localization technology of wireless sensor network and the principle of several common localization algorithms,and gives the performance index of localization algorithm.Finally,several localization algorithms are compared,and the influence of node distribution on localization results is analyzed.(3)This thesis analyzes the shortcomings of APIT localization algorithm,and summarizes the reasons for the shortcomings,and then proposes a hybrid localization algorithm based on APIT.The hybrid algorithm optimizes the grid scanning phase of APIT localization algorithm through triangle comparison and segmentation,and combines with genetic algorithm to locate the blind node with only two reference nodes in the communication range and improve the localization accuracy and coverage.(4)The localization system of wireless sensor network is designed and implemented by CC2530 chip and IAR embedded workbench software development environment.The Z-stack protocol stack is transplanted in CC2530,and the coordinator,reference node and blind node are used to form the topology network.The collected data is brought into the algorithm implementation,and the blind node position is displayed through the GUI of MATLAB.In order to verify the effectiveness of hybrid algorithm,the software simulation and hardware experiment are carried out under the same experimental environment.The results show that the hardware experiment and software simulation results are consistent in the whole,and the error is controlled within a reasonable range.This shows that the hybrid algorithm is not only theoretically feasible,but also has certain practical application value.更多还原