恶意节点攻击下的基于RSS测量值的压缩感知定位算法

发布时间：2018-06-24 12:58

本文选题：安全定位算法 + 接收信号强度　；参考：《南京邮电大学》2017年硕士论文

【摘要】：在传统的无线传感器网络(WSN)定位技术中,基于接收信号强度(RSS)的定位技术实现简单,受到学术界和工业界的高度关注。将压缩感知理论应用到WSN定位中可以有效减少RSS测量值数目,同时具有更好的定位性能。但是在实际应用中RSS测量值很容易受到物理攻击。因此本文以基于RSS的压缩感知定位系统为背景,研究恶意节点攻击下的安全定位算法,主要工作包括:(1)总结了基于压缩感知的定位算法的国内外研究现状,详细介绍了压缩感知相关知识;建立了基于压缩感知的网络定位模型,分析了该定位模型的安全性,说明了其安全隐患的由来;并建立了模拟RSS攻击的线性攻击模型,为安全定位机制提供理论支撑。(2)提出了基于平方中位数法的压缩感知安全定位算法。该算法利用残差最小中位数来分析检测受攻击节点,有效保证了定位系统的准确性。同时,本文结合理论分析和实验仿真,解决了算法参数设计的问题,使用更合理的参数来确保算法发挥最大的抗攻击性能。(3)提出了基于残差分析的压缩感知安全定位算法。首先利用压缩感知重构算法估计目标位置,并计算本次估计的目标位置对应的残差向量;然后,利用残差偏差准则,找出所有可疑的受攻击节点集合,同时根据最大残差准则确定受攻击的测量值;最后,当残差和趋于稳定,说明所有受攻击的节点已经被删除,利用未受攻击的测量值进行最终位置估计。仿真实验表明,在存在恶意节点攻击的条件下,本算法能保持很好的定位性能。
[Abstract]:In the traditional wireless sensor network (WSN) localization technology, the location technology based on received signal strength (RSS) is easy to implement, and has been highly concerned by academia and industry. Applying compression sensing theory to WSN localization can effectively reduce the number of RSS measurements and have better localization performance. However, RSS measurement is vulnerable to physical attack in practical applications. Therefore, based on the RSS based compressed sensing localization system, this paper studies the security localization algorithm under malicious node attack. The main work includes: (1) summarize the research status of compressed sensing localization algorithm at home and abroad. The related knowledge of compression perception is introduced in detail, the network location model based on compression perception is established, the security of the location model is analyzed, and the origin of its security hidden danger is explained, and the linear attack model of simulating RSS attack is established. It provides theoretical support for the security localization mechanism. (2) A compressed sensing security location algorithm based on the square median method is proposed. The algorithm uses the minimum median of residual error to analyze and detect the attacked nodes, which effectively ensures the accuracy of the location system. At the same time, combined with theoretical analysis and experimental simulation, this paper solves the problem of algorithm parameter design, using more reasonable parameters to ensure the maximum anti-attack performance of the algorithm. (3) A compressed sensing security location algorithm based on residual analysis is proposed. Firstly, the target location is estimated by the compressed perceptual reconstruction algorithm, and the residual vector corresponding to the target position is calculated. Then, using the residual deviation criterion, the set of all the suspected attacking nodes is found. At the same time, according to the maximum residual error criterion, the measured value of the attack is determined. Finally, when the residual value is stable, it shows that all the attacked nodes have been deleted, and the final position estimation is carried out by using the unattacked measurement value. Simulation results show that the algorithm can maintain good localization performance under the condition of malicious node attack.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5;TP212.9

【参考文献】