Statistical analysis of single-frequency GNSS positioning error under a suspected spoofing attack

This paper investigates the statistical behaviour of single-frequency GNSS positioning errors under conditions of a suspected spoofing attack. Data from the IGS reference station in Nicosia, Cyprus, were analysed by comparing two cases: a period of significant GNSS PNT performance degradation potentially associated with spoofing and a control period representing usual GNSS performance. Positioning error components for individual GNSS systems were derived from raw GNSS pseudorange observations, with solutions based on both broadcast and received navigation messages. Position estimates were obtained using RTKLIB/RTKPOST, while exploratory and statistical analyses were performed using software developed by the authors in the open-source R environment for statistical computing. The analysis included statistical properties of positioning errors, comparative graphical analysis, correlation analysis, and statistical hypothesis testing. The results did not provide evidence of spoofing caused by manipulation of navigation messages; however, they indicated degradation of GNSS positioning performance with different levels of exposure across the analysed GNSS systems. The main contributions of the research include a methodology for detecting false-navigation-message-based GNSS spoofing, a procedure for scenario selection and assessment, and a tailored software research framework developed in R for statistical computing.