نوع مقاله : مقاله پژوهشی
نویسندگان
1 دانشجوی کارشناسی، دانشگاه علوم و فنون هوایی شهید ستاری، تهران، ایران
2 استادیار، دانشگاه علوم و فنون هوایی شهید ستاری، تهران، ایران
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
In the past decade, the protection of GNSS satellite systems against spoofing and jamming attacks has become an important focus. In a spoofing attack, the GNSS signal receiver makes errors in determining location and time because the spoofed signal is tracked by the receiver due to its close resemblance to the authentic signal and its higher power. Currently, there is no comprehensive or universal method for detecting all types of spoofing attacks. In this paper, a novel approach for detecting spoofing signals using unsupervised machine learning algorithms is presented. Two machine learning algorithms, including the density-based clustering algorithm OPTICS and the Gaussian Mixture Model (GMM), are proposed to detect spoofing attacks. These algorithms are used to distinguish between genuine GPS signals and spoofed or fake signals. The original and spoofed signals have differences in features such as phase variance, correlation distribution and signal energy, which form the basis for clustering. The performance of these algorithms has been evaluated using Silhouette scores and the confusion matrix. In addition, the algorithms were implemented and tested on a GPS software receiver. Spoofing attacks were successfully detected with an accuracy of 92.45% and 99.88% respectively.
کلیدواژهها [English]
)