无线传感器网络定位算法及其安全研究

发布时间：2018-03-28 14:32

本文选题：无线传感器网络　切入点：定位求精　出处：《杭州电子科技大学》2017年硕士论文

【摘要】：无线传感器网络(WSNs)在军事、工农业控制、环境监测等相关领域有着广阔的应用前景,作为基于WSNs的应用的基础,节点定位技术是WSNs中极其重要的技术之一。一方面,精确的位置信息对于WSNs的应用服务具有非常关键的作用;另一方面,由于传感器节点的资源非常有限,包括能量、存储容量等,而且节点部署的环境常常较为恶劣,所以在WSNs的定位过程中,能耗和安全性问题也是不容忽视。本文以经典的无需测距DV-hop定位算法为研究对象,从定位精度,能耗和安全性三个方面展开研究,主要贡献如下:(1)提出了一种基于校正矢量的分布式迭代求精算法(CVLR)。为了解决DV-hop算法中的定位模糊问题,CVLR算法提出利用节点的一跳或两跳邻居节点信息构建位置校正矢量,然后采用搜索算法对DV-hop定位结果进行迭代求精,最终实现CVLR1(仅仅利用一跳邻居节点信息)和CVLR2(同时利用一跳和两跳邻居节点信息)两种算法。实验表明,CVLR1和CVLR2的定位精度明显比原始DV-hop算法高,从实验数据看,两种算法的定位精度分别相对于DV-hop算法提高了41.12%和56.61%。(2)对DV-hop、CVLR1和CVLR2算法的能耗进行了分析,在此基础上,为了结合CVLR2的高精度和CVLR1的低能耗,我们提出了基于精度-能耗联合优化的改进CVLR算法(ECVLR)。该算法主要是减少了CVLR2在迭代过程中节点间互相交换信息的通信开销,使其在既能同时利用一跳和两跳邻居节点信息进行求精的情况下,又能减少能耗。仿真实验表明,ECVLR定位精度相对于CVLR1提高了21.37%,而能耗相对于CVLR2下降了86.69%,该算法在定位精度和能耗两者之间做了很好地平衡,是一种具有较高能效性的定位算法。(3)提出了DV-hop定位中普通节点被俘获的欺骗攻击模型,分析了这种欺骗攻击模型对DV-hop定位过程的影响,进而提出了一种抵御欺骗攻击的DV-hop安全定位算法。该算法在普通节点端利用发送-转发信息一致性来检测恶意节点,在汇聚节点端利用消息转发链举证来确定恶意节点,当汇聚节点检测出存在恶意节点进行篡改攻汇聚节点通报全网弃用恶意节点转发的数据分组并重启定位。仿真实验表明,本文提出的安全定位算法能够有效滤除恶意节点,且安全定位算法的定位性能与无攻击下的DV-hop算法基本相当,在仅有1个恶意节点的条件下,安全定位算法的定位精度相对于无攻击下的DV-hop仅下降0.79%,这说明我们的安全定位算法有效的消除了欺骗攻击对DV-hop造成的影响。
[Abstract]:Wireless sensor network (WSNs) has a broad application prospect in the military, industrial and agricultural control, environmental monitoring and other related fields. As the basis of the application based on WSNs, node location technology is one of the most important technologies in WSNs. Accurate location information plays a key role in WSNs application services. On the other hand, because of the limited resources of sensor nodes, including energy, storage capacity and so on, nodes are often deployed in a bad environment. Therefore, the problem of energy consumption and security is not to be ignored in the positioning process of WSNs. This paper takes the classical location algorithm without ranging DV-hop as the research object, from three aspects of location accuracy, energy consumption and security. The main contributions are as follows: (1) A distributed iterative refinement algorithm based on correction vector is proposed. In order to solve the problem of location ambiguity in DV-hop algorithm, a position correction vector is constructed by using one-hop or two-hop neighbor node information of nodes. Then the search algorithm is used to iterate and refine the DV-hop localization results. Finally, two algorithms, CVLR1 (using only one-hop neighbor node information) and CVLR2 (using one-hop and two-hop neighbor node information), are realized. The experimental results show that the positioning accuracy of CVLR1 and CVLR2 is obviously higher than that of the original DV-hop algorithm. The positioning accuracy of the two algorithms is improved by 41.12% and 56.61% respectively compared with the DV-hop algorithm. The energy consumption of the DV-hopper CVLR1 and CVLR2 algorithm is analyzed. On this basis, in order to combine the high precision of CVLR2 with the low energy consumption of CVLR1, We propose an improved CVLR algorithm based on the combined optimization of precision and energy consumption, which can reduce the communication overhead of exchanging information between nodes in the iterative process of CVLR2. So that it can make use of both one-hop and two-hop neighbor node information for refinement, The simulation results show that the positioning accuracy of CVLR1 is 21.37% higher than that of CVLR1, while the energy consumption is 86.69% lower than that of CVLR2. This algorithm has a good balance between positioning accuracy and energy consumption. This paper presents a spoofing attack model of common nodes captured in DV-hop location, and analyzes the influence of this spoofing attack model on DV-hop localization process. Furthermore, a DV-hop security localization algorithm against spoofing attacks is proposed, which detects malicious nodes by sending-forward message consistency at the common node end, and determines malicious nodes by using message forwarding chain at the convergence node side. When the convergence node detects that there are malicious nodes tampering with the convergence node to notify the whole network to discard the malicious node forwarding data packet and restart the location. The simulation results show that the proposed security location algorithm can effectively filter out the malicious node. And the location performance of the secure location algorithm is basically the same as that of the DV-hop algorithm without attack, under the condition of only one malicious node, The location accuracy of the secure location algorithm is only 0.79% lower than that of the DV-hop without attack, which shows that our security location algorithm effectively eliminates the influence of spoofing attack on DV-hop.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【参考文献】