Auto-navigating of unmanned aerial vehicles (UAV) in the outdoor environment is performed by using the Global positioning system (GPS) receiver. The power of the GPS signal on the earth surface is very low. This can affect the performance of GPS receivers in the environments contaminated with the other source of radio frequency interference (RFI). GPS jamming and spoofing are the most serious and intentional RFI attacks. Due to the jamming attacks, the positioning accuracy of UAV will be degraded. This positioning error can be reached to tens of kilometers off the true location as was reported in some research articles. The detection of GPS jamming is the first step to mitigate this attack. Most of jamming detection methods are based on GPS signals processing. If a jamming detection system just relies on GPS signal processing to detect and confront RFI threats, it will be capable of failure in the use of the more advanced hardware and sophisticated methods by invaders intentional. The camera is a common sensor almost in all of UAVs which is a passive sensor and resistant to the jamming signals. Here, the use of image-based navigation methods for GPS jamming detection will be helpful due to the insensitivity of camera sensors to jamming signals. GPS jamming attacks can be detected in UAV navigation, independently of the signal processing methods, using a comparison of two flight trajectories assigned for a UAV, determined from visual navigation and GPS positioning data. For this purpose, the trajectory descriptor of the Normalized Distance of the Consecutive Points (NDCP) and trajectory descriptor of the Consecutive Directions Angles (CDA) is used. These descriptors are independent of the coordinate system of the trajectory but are not independent of the number of trajectory points. As a result of the jamming attacks, the GPS receiver may not be able to receive the signal and the positioning cannot be performed. Therefore, two trajectories of UAV from GPS and visual navigation may have a different number of points. So, the NDCP and CDA cannot be used for these trajectories in all the time. The trajectory descriptor of Histogram of the Oriented Displacements (HOD) is independent of the number of trajectory points, but it is not independent of the coordinate system of the trajectory. In this paper, a method is developed to allow the use of the HOD trajectory descriptor to detect the occurrence of the GPS jamming attacks. For this purpose, first, the coordinate system of the UAV trajectory from GPS data is transformed to the coordinate system of the UAV trajectory from visual navigation. Here, a sliding window-based approach is used for determining of the jamming location. The performance of the HOD trajectory descriptor in detecting jamming attacks versus the NDCP and CDA trajectory descriptors were compared. The results show that the HOD trajectory descriptor has a significant advantage in detecting the jamming attacks concerning NDCP and CDA trajectory descriptors, especially in the positioning errors of more than ten meters due to jamming attacks, which can be more reliable. This descriptor can be used to detect the occurrence of the jamming attacks. The ability of the HOD trajectory descriptor is significant in detecting positioning errors greater than five meters, compared to the other two trajectory descriptors.
GPS Jamming Detection in UAV Navigation Using Visual Odometry and HOD Trajectory Descriptor
Journal of Geomatics Science and Technology, Volume:9 Issue:1, 2019
15 - 30
روشهای دسترسی به متن این مطلب
در سایت عضو شوید و هزینه اشتراک یکساله سایت به مبلغ 300,000ريال را پرداخت کنید. همزمان با برقراری دوره اشتراک بسته دانلود 100 مطلب نیز برای شما فعال خواهد شد!
به کتابخانه دانشگاه یا محل کار خود پیشنهاد کنید تا اشتراک سازمانی این پایگاه را برای دسترسی همه کاربران به متن مطالب خریداری نمایند!