Detecting and Localizing Space Based Interference on GNSS Signals Using Machine Learning

Global Navigation Satellite Systems (GNSS) heavily depend on low-power signals, which operate below the noise floor. This makes them susceptible to interference that can significantly disrupt navigation. Such interference can degrade a receiver's accuracy and reliability in generating Position Navigation and Timing (PNT) solutions and even prevent it from acquiring and tracking nearby GNSS signals. Ground-based interference is relatively easy to identify with a single receiver, its sources can be diverse and widespread. Space-based interference, on the other hand, is difficult to detect without an extensive receiver network, followed by additional complications in identifying its source. One instance of a space-based interference that began in June 2021 led to the investigation and detection of an unusual power spike in the B3/E6 band, centered at 1268.52 MHz using Trimble's global network of 43 multifrequency receivers. This network is spread across the US and Europe as described in Patil et al., (2023). The interference event, exhibiting a distinct pattern, was confirmed to be space-based due to its simultaneous impact on the entire receiver network for over 24 hours. This paper extends that previous work by further characterizing the interference on B3I, modeling the potential effects of different interferers on various GNSS frequencies, and developing a method for streamlining the process of detecting a space-based interference using machine learning. The goal is to enhance the capability of widely-distributed receiver networks in promptly detecting and attributing the source of interference, aiding in the improvement of GNSS reliability.