【作者】 孙庭波；

【导师】 屈玉贵；

【作者基本信息】 中国科学技术大学 ， 通信与信息系统， 2008， 博士

【摘要】 在无线传感器网络中,确定节点位置或事件发生的位置对其监测活动至关重要。节点自身的准确定位不仅是提供监测事件或监测目标位置信息的前提,也是提供网络拓扑自配置、提高路由效率、向部署者报告网络的覆盖质量以及为网络提供命名空间等网络功能的基础。传感器网络固有的资源、能量限制使节点定位变得十分困难,传感器网络的多样性也导致不存在适用于各种网络环境的通用定位算法。现有无线传感器网络定位算法一般适用于某种特定的节点硬件和网络应用。因此,针对特定的网络环境设计和优化相应的定位算法,降低对网络同步、能量消耗等方面的要求,对于传感器网络技术的发展和实用性的提高具有重要的价值。绝大多数传感器网络节点定位系统均假定具有安全可信的网络环境,忽略了定位过程中的安全问题。然而传感器网络的开放性和无人看护性使节点的定位过程极易受到来自外部敌对者或内部被劫持节点的攻击。对节点定位系统的攻击往往具有明确的针对性,同时攻击手段种类较多、因系统所采用的定位技术和过程不同而有所不同。一般的安全措施难以抵御这种具有针对性和多样性的安全威胁。攻击所造成的无效或错误的定位结果可能给传感器网络应用,尤其是那些具有重要使命的应用(例如战场监视),造成难以估量的重大损失。因此,如何为存在敌对可能的传感器网络提供尽量安全的节点定位系统,是一个必须解决的关键问题。本文主要工作和创新之处:●提出了分段式异步网络定位算法在分析和比较已有的传感器网络定位算法基础上提出了一种用于异步无线传感器网络定位运动目标的分段式定位算法。该算法根据目标移动速度将定位过程自动分段,在每个时间段初期用同步定位算法获得目标准确初始位置,然后在此基础上采用异步定位。仿真结果表明分段式定位算法大大降低了网络同步开销,解决了普通异步定位算法无法准确定位慢速目标的问题,具有实用性。●提出了SAPIT安全定位算法对不同定位技术所面临的攻击种类进行了分析,深入研究用于安全定位的措施的实现原理、特点、局限和彼此的联系,并提出了一种新的无线传感器网络安全定位算法SAPIT:它采用锚节点身份验证、可疑锚节点检测、APIT测试结果安全汇总等安全机制。在不增加高精度定时器、方向天线等硬件的条件下,大大增强定位算法的安全性。对该算法模拟仿真,结果表明SATPIT能有效抵御干扰、虫洞攻击、劫持节点进行的伪装锚节点攻击和InToOut等攻击。●搭建了传感器网络节点试验环境研制了自主CPU的传感器网络节点,并在此平台上验证了SAPIT算法的实用性。更多还原

【Abstract】 In wireless sensor networks, the location of the nodes and monitored events is of critical importance to their monitoring activities. Accurate localizaion of nodes is not only the premise of monitoring events or targets, but also the foundation of network applications such as reporting network topology, improving the efficiency of routing, reporting the deployment of network coverage or quality and providing network name space. Location detection is complicated in sensor networks which are characterized by limited resources. No localization schemes can be used for all kinds of sensor networks. Existing positioning methods are only suite for a particular nodal hardware or network environment. Designing special positioning methods for specific nodes and working environments is very important for the practicability of sensor network technology.Most existing positioning systems assume secure and truthful environments and ignore security issues in the process of positioning. However sensor networks often work in open and unattended environments and are vulnerable to suffer from internal attacks as well as the external ones. Attacks against positioning system are often pointed and vary with different localization methods and processes. The invalid or wrong positioning results caused by attackers will bring incalculable losses to sensor network applications, especially to those with an important mission (for example, battlefield surveillance). Therefore, it’s essential to design a secure positioning system for WSN working in hostile environments.The major work and new ideas of this dissertation are as follows:Firstly, it focuses on analyzing and comparing existing location algorithms and then a segmented localization scheme is proposed for location system using asynchronous distributed wireless sensor networks. This scheme segments the localization process according to the source speed. At the beginning of each segment, synchronous localization is performed to get a good estimation of source location and asynchronous localization is started after that. Segmented localization scheme, relaxing the assumption of source motion speed, solves the problem that ordinary asynchronous localizations cannot work well with slowly moving source.Secondly, it analyzes attacks against location system and various secure methods. Also, a new range-free secure localization scheme, called SAPIT, is presented. SAPIT utilizes anchor ID authentication and APIT test results safe aggregation. It solves the problem that APIT is easily attacked by compromised nodes, without any extra hardware. Simulation results indicates that SAPIT can resist jamming, wormhole, false anchors and InToOut attacks.At last, it provides a hardware platform and SAPIT is implemented in this platform.更多还原