基于选举策略的压缩感知定位算法

发布时间：2018-06-25 22:51

本文选题：无线传感器网络 + 压缩感知　；参考：《南京邮电大学》2017年硕士论文

【摘要】：随着无线通信网络技术的不断发展,基于位置的服务越来越受到人们的关注,基于接收信号强度(Received Signal Strength,RSS)的定位方法因原理简单、成本低,广泛应用于工业领域。但是该类算法易于遭受恶意节点的攻击,导致定位效果不佳。基于上述问题,本文利用压缩感知技术,研究了基于选举策略的压缩感知定位算法。主要研究内容如下:(1)详细阐述了无线传感器网络定位方法。详细描述了压缩感知基本理论,对基追踪(Basic Pursuit,BP)、正交匹配追踪(Orthogonal Matching Pursuit,OMP)、匹配追踪(Matching Pursuit,MP)、压缩采样匹配追踪(Compressive Sampling Matching Pursuit,CoSaMP)、子空间追踪(Subspace Pursuit,SP)等几种常见的压缩感知重构算法进行了仿真实验,并且分析了每种算法的重构性能。(2)提出了基于选举策略和OMP重构技术的安全定位算法。首先利用一种稳健算法产生可靠的校验参数估计,然后利用投票策略来提高在异常值存在时的稳健性。仿真结果说明,在恶意节点攻击下,该算法可以实现比其他现有算法更好的定位性能。(3)将正交匹配-子空间追踪(OMP-SP)重构技术用于无线传感器网络定位,提出了基于OMP-SP重构技术的定位算法。在恶意节点攻击的环境下,提出了基于选举策略和OMP-SP重构技术的安全定位算法。仿真结果表明,基于OMP-SP的定位误差要小于基于OMP和基于SP算法的定位误差。在恶意节点攻击环境下,安全定位算法能够克服恶意节点攻击,提高定位精度。
[Abstract]:With the development of wireless communication network technology, location-based service has attracted more and more attention. The location method based on received signal strength (RSS) is widely used in industrial field because of its simple principle and low cost. However, this kind of algorithm is vulnerable to malicious node attack, resulting in poor localization effect. Based on the above problems, this paper studies the localization algorithm of compressed perception based on election strategy using compressed sensing technology. The main research contents are as follows: (1) the location method of wireless sensor network is described in detail. The basic theory of compressed sensing is described in detail. Simulation experiments are carried out on several common compression perception reconstruction algorithms, such as basic pursuit BP, orthogonal matching pursuit OMP, matching pursuit MP, Compressive sampling matching pursuit CoSaMP, subspace pursuit SP, and so on. The reconstruction performance of each algorithm is analyzed. (2) A secure location algorithm based on election strategy and OMP reconstruction is proposed. First, a robust algorithm is used to generate reliable checkout parameter estimation, and then the voting strategy is used to improve the robustness when outliers exist. Simulation results show that the algorithm can achieve better localization performance than other existing algorithms under malicious node attack. (3) orthogonal matching subspace tracking (OMP-SP) reconstruction technology is used to locate wireless sensor networks. A localization algorithm based on OMP-SP reconstruction technology is proposed. In the environment of malicious node attack, a secure location algorithm based on election strategy and OMP-SP reconstruction technology is proposed. Simulation results show that the location error based on OMP-SP is smaller than that based on OMP and SP algorithm. In the malicious node attack environment, the security localization algorithm can overcome the malicious node attack and improve the location accuracy.
【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】