Feature Selection for GNSS Receiver Fingerprinting

Fingerprinting is the process of gathering information about an electronic device to generate specific signatures to identify the device itself. This working paper investigates potential strategies for fingerprinting a GNSS receiver. Several metrics computed from the time series provided by the receiver are analysed and considered as candidate features for fingerprinting. It is shown that features computed from the receiver clock drift are the most appropriated to determine the receiver signature. Three clock-derived metrics are sufficient to determine a receiver fingerprint and build a “white list” for receiver identification.