基于粒子滤波的GNSS抗欺骗式干扰研究
发布时间：2018-08-17 15:09
【摘要】：全球导航卫星系统(Global Navigation Satellite System，GNSS)可以在全球范围内提供全天候的定位、测速和授时业务。然而在其被广泛应用的同时，其潜在的风险也引起了关注。欺骗式干扰攻击通过转发或者再生的方式伪造的干扰信号，误导接收机定位授时。本课题针对GNSS的欺骗式干扰，提出基于分块粒子滤波解算的改进型RAIM(Receiver Autonomous Integrity Monitoring)抗干扰方案，其在提高定位授时精度的同时，还可以有效地消除干扰信号的影响。 首先，本文对GNSS接收机解算的基本原理和最小二乘解算的理论基础进行介绍。接收机根据测得的伪距和伪距变化率，以及从星历中获知的卫星位置信息和速度信息，可以建立一个关于自身位置和速度信息的非线性方程组，用以定位和测速解算。经典的最小二乘解算算法借助牛顿迭代使非线性方程组线性化，然后利用最小二乘法实现对位置与速度的估计。 为了提高GNSS接收机的定位解算精度，本文将粒子滤波引入到导航解算流程，并根据实际情况提出一种可以进一步提高解算精度的分块粒子滤波解算算法。粒子滤波借助蒙特卡洛模拟实现对期望变量概率密度函数的近似估计，从而实现递推贝叶斯滤波，是一种非参数化的近似最优非线性滤波方法。基于粒子滤波的解算算法，可以提供比最小二乘法更高的定位、测速和授时精度。而后本文提出的分块粒子滤波解算算法，将传统粒子滤波算法的位置估计和速度估计分离，降低状态矢量和粒子的维度，如此可以在粒子数量不变的条件下，，使当前粒子更好地覆盖状态矢量空间，进而提高解算精度。 最后，针对GNSS面临的欺骗式干扰威胁，提出一种基于分块粒子滤波解算的改进型RAIM抗干扰方案。无论是转发式还是产生式欺骗干扰，其能够误导定位的主要原因都是干扰信号所携带的附加伪距；根据附加伪距的不同特征，可以将干扰场景分为伪距缓变场景和伪距跳变场景，每种场景下又包含单星干扰和多星干扰两种情况。通过分析不同场景、不同干扰卫星数量和不同模式的干扰攻击对观测矢量和定位误差的影响，提出一种改进型RAIM的抗欺骗式干扰方案。该方案利用观测矢量的模值进行干扰攻击检测，同时利用观测矢量和残余矢量之间的线性相关性实现干扰卫星的识别。从仿真结果可知，本文提出的改进型RAIM可以有效地检测和识别欺骗式干扰，而包含有改进型RAIM的分块粒子滤波解算算法可以在干扰攻击时，依然输出正确的定位结果，这也证明了本文提出的GNSS抗欺骗式干扰算法的有效性。
[Abstract]:GNSS (Global Navigation Satellite system (Global Navigation Satellite) can provide all-weather positioning, velocimetry and timing services around the world. However, while it is widely used, its potential risks have attracted much attention. Spoofed jamming attacks mislead the receiver by forgery of jamming signals by forwarding or reproducing them. Aiming at the deceptive interference of GNSS, an improved RAIM (Receiver Autonomous Integrity Monitoring) anti-jamming scheme based on block particle filter solution is proposed in this paper. It can improve the accuracy of location timing and eliminate the influence of interference signal at the same time. Firstly, the basic principle of GNSS receiver and the theoretical basis of least square solution are introduced. According to the measured pseudo-range and pseudo-range rate, and the satellite position information and velocity information obtained from the ephemeris, the receiver can establish a nonlinear system of equations about its position and velocity information, which can be used to locate and calculate the velocity measurement. The classical least square algorithm linearizes the nonlinear equations by Newton iteration, and then estimates the position and velocity by using the least square method. In order to improve the positioning accuracy of GNSS receiver, the particle filter is introduced into the navigation solution flow, and a block particle filter algorithm which can further improve the accuracy of the algorithm is proposed according to the actual situation. Particle filter is a non-parameterized approximate optimal nonlinear filtering method by using Monte Carlo simulation to realize approximate estimation of probability density function of expected variables and to realize recursive Bayesian filtering. The algorithm based on particle filter can provide higher accuracy of localization, velocity measurement and timing than the least square method. Then the partitioned particle filter algorithm proposed in this paper separates the position estimation from the velocity estimation of the traditional particle filter algorithm to reduce the state vector and particle dimension, so that the number of particles can not be changed. The state vector space is better covered by the current particle, and the accuracy of the solution is improved. Finally, an improved RAIM anti-jamming scheme based on block particle filter is proposed for the spoofed jamming threat faced by GNSS. Whether forwarding or producing spoofing interference, the main reason why it can mislead the location is the additional pseudo-range carried by the jamming signal; according to the different characteristics of the additional pseudo-range, The jamming scene can be divided into pseudo-range slow-varying scene and pseudo-range jump scene, and each scenario includes two kinds of single-star interference and multi-star interference. By analyzing the effects of different scenes, different number of jamming satellites and different modes of jamming attack on the observation vector and positioning error, an improved anti-deception jamming scheme of RAIM is proposed. In this scheme, the interference attack is detected by the mode value of the observation vector, and the linear correlation between the observation vector and the residual vector is used to identify the jamming satellite. The simulation results show that the proposed improved RAIM can effectively detect and identify spoofed interference, while the block particle filter algorithm with improved RAIM can still output correct localization results during interference attack. This also proves the effectiveness of the GNSS anti-deception algorithm proposed in this paper.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN96.1
本文编号：2188035
[Abstract]:GNSS (Global Navigation Satellite system (Global Navigation Satellite) can provide all-weather positioning, velocimetry and timing services around the world. However, while it is widely used, its potential risks have attracted much attention. Spoofed jamming attacks mislead the receiver by forgery of jamming signals by forwarding or reproducing them. Aiming at the deceptive interference of GNSS, an improved RAIM (Receiver Autonomous Integrity Monitoring) anti-jamming scheme based on block particle filter solution is proposed in this paper. It can improve the accuracy of location timing and eliminate the influence of interference signal at the same time. Firstly, the basic principle of GNSS receiver and the theoretical basis of least square solution are introduced. According to the measured pseudo-range and pseudo-range rate, and the satellite position information and velocity information obtained from the ephemeris, the receiver can establish a nonlinear system of equations about its position and velocity information, which can be used to locate and calculate the velocity measurement. The classical least square algorithm linearizes the nonlinear equations by Newton iteration, and then estimates the position and velocity by using the least square method. In order to improve the positioning accuracy of GNSS receiver, the particle filter is introduced into the navigation solution flow, and a block particle filter algorithm which can further improve the accuracy of the algorithm is proposed according to the actual situation. Particle filter is a non-parameterized approximate optimal nonlinear filtering method by using Monte Carlo simulation to realize approximate estimation of probability density function of expected variables and to realize recursive Bayesian filtering. The algorithm based on particle filter can provide higher accuracy of localization, velocity measurement and timing than the least square method. Then the partitioned particle filter algorithm proposed in this paper separates the position estimation from the velocity estimation of the traditional particle filter algorithm to reduce the state vector and particle dimension, so that the number of particles can not be changed. The state vector space is better covered by the current particle, and the accuracy of the solution is improved. Finally, an improved RAIM anti-jamming scheme based on block particle filter is proposed for the spoofed jamming threat faced by GNSS. Whether forwarding or producing spoofing interference, the main reason why it can mislead the location is the additional pseudo-range carried by the jamming signal; according to the different characteristics of the additional pseudo-range, The jamming scene can be divided into pseudo-range slow-varying scene and pseudo-range jump scene, and each scenario includes two kinds of single-star interference and multi-star interference. By analyzing the effects of different scenes, different number of jamming satellites and different modes of jamming attack on the observation vector and positioning error, an improved anti-deception jamming scheme of RAIM is proposed. In this scheme, the interference attack is detected by the mode value of the observation vector, and the linear correlation between the observation vector and the residual vector is used to identify the jamming satellite. The simulation results show that the proposed improved RAIM can effectively detect and identify spoofed interference, while the block particle filter algorithm with improved RAIM can still output correct localization results during interference attack. This also proves the effectiveness of the GNSS anti-deception algorithm proposed in this paper.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN96.1
【参考文献】
相关期刊论文 前2条
1 卢丹;吴仁彪;王磊;;一种通用的GPS多类干扰抑制方法[J];信号处理;2010年05期
2 李四海;刘洋;张会锁;张晓冬;;惯性信息辅助的卫星导航欺骗检测技术[J];中国惯性技术学报;2013年03期
本文编号：2188035
本文链接：https://www.wllwen.com/kejilunwen/wltx/2188035.html
