GNSS spoofing detection using a fuzzy classifier based on time–frequency analysis of the autocorrelation function

• Published in: GPS solutions 2024-07, Vol.28 (3), p.146, Article 146
• Main Authors: Tohidi, S., Mosavi, M. R.
• Format: Article
• Language: English
• Subjects: Algorithms; Antennas; Artificial Intelligence Applications in GNSS; Atmospheric Sciences; Autocorrelation functions; Automotive Engineering; Classifiers; Digital signatures; Discrete Wavelet Transform; Earth and Environmental Science; Earth Sciences; Electrical Engineering; Frequency analysis; Geophysics/Geodesy; Global navigation satellite system; Methods; Original Article; Receivers & amplifiers; Satellites; Sensors; Signal processing; Space Exploration and Astronautics; Space Sciences (including Extraterrestrial Physics; Spoofing; Time-frequency analysis; Wavelet transforms
• ISSN: 1080-5370; 1521-1886
• DOI: 10.1007/s10291-024-01674-y

We present a method to detect spoofing attack in Global Navigation Satellite System signals for single antenna receivers based on autocorrelation function distortion analysis in the Time–Frequency (TF) domain. In particular, Discrete Wavelet Transform (DWT) is considered as a TF tool to investigate the correlation taps outputs of the received signal. The statistical properties of the DWT coefficients of the autocorrelation function are processed in a fuzzy classifier as a feature vector to discriminate the presence of a spoofing attack. The detection performance of the method based on TF analysis of the autocorrelation function is verified using the real well-known Texas Spoofing Test Battery (TEXBAT) dataset. The findings demonstrate that the suggested technique for P fa  = 10 −2 yields an average detection rate of more than 95% for the TEXBAT different cases, which shows improved detection sensitivity and robustness compared to other conventional and state-of-art methods.