Impact of RFI on GNSS and Avionics – A View from the Cockpit

Abstract

Introduction The increase of Radio Frequency Interference (RFI) events is a well-documented trend, that is becoming a threat to the use of Global Navigation Satellite Systems (GNSS) in aviation. EUROCONTROL’s voluntary ATM incident reporting system EVAIR is a platform for pilot to report relevant incidents, such as GNSS outages. In the years 2018 and 2019 more than 4300 and 3500 GPS outages were reported compared to less than 500 events in the previous years. While the tool is giving a good geographic perspective about the typical RFI hotspots, it is ultimately based on voluntarily reported data. As pilots get used to widespread RFI in certain areas and Notices to Airmen (NOTAMs) are published that RFI should be expected, these events are considered to be normal. This likely leads to the numbers behind the GNSS outages reported in EVAIR being significantly underestimated. In this presentation we will give the audience a feel for why we believe this to be true: outages in GNSS service due to RFI initially manifest themselves in the cockpit in subtle ways, not immediately identifiable to the unaware. Background The Nicosia Flight Information Region (FIR) has long reported difficulties with GNSS service coverage. It regularly issues NOTAMs warning pilots of the phenomenon and kindly requests radio reports of such observations. Pilots regularly report loss of GNSS on approach to Larnaca International Airport (LCA) making the introduction of GNSS-based approach procedures impossible. While many aircraft can cope with GNSS outages easily by using terrestrial navigation aids and/or Inertial Reference Systems onboard the aircraft, some aircraft do not have this option and may require ATC support. This is causing an increase in workload for air traffic controllers and pilots of the affected aircraft alike. As a way to better understand the prevalence of the phenomenon and its implications on main-line aircraft, DLR flew its Airbus 320 to LCA, equipped with a variety of GNSS monitoring instruments, including the baseline avionics of the aircraft, a set of experimental GNSS receivers and a recorder for sampling the raw radio signal. Observations and Data An important feature of the flight is that we have been able to analyze the phenomenon from several perspectives. This includes photographs of the cockpit displays, and matching recordings of GNSS signals, where available, in terms of pseudorange and signal-to-noise ratio. In addition, we collected raw, pre-correlator signal samples, the analysis of which offers a very particular type of insight. The raw signal was captured with a recording device, after being downconverted to baseband. The sampling rate was set to 30.7 MHz, at eight bits per sample. This quality of recording allows us to study the response of a GNSS receiver in “playback” mode, but it also enables an analysis of the frequency contents of the radio band in question. Reception and Impact While it is not a particularly novel message that GNSS signals are being jammed with RFI in some parts of the world, the aviation community has already reacted to our findings. Some of the results have been discussed at EUROCONTROL-hosted events, as well as a white paper. The Cypriot Department of Civil Aviation has, since, stated that the measurements prompted a revision of the threat model behind the RFI events. As for the wider aviation community, these observations and measurements clearly show a need to revise the paradigms for the provision of resilient navigation services. An outdate understanding of the RFI threat was based on individual drivers sporadically disrupting GNSS services at airports, for example by using GNSS privacy devices. Our measurements make it evident that this threat model is outdated, as GNSS outages clearly last several hours (potentially even more) and cover an entire FIR. The Presentation We will present the measurements and results discussed above in a presentation designed to create awareness of the impact of RFI on avionics hardware. During this flight we observed that it is easy for pilots to underappreciate the degree of disruption to their GNSS avionics. An important part of the message is that we have found no evidence of intentional misleading, inside the GNSS signals. The service disruptions are likely collateral to the geopolitical reality of the region. In this sense, the results show that there is no direct threat to the safety of air traffic and, consequently, the need to act may not be acute. The results do, however, show a need to modernize the airspace and CNS infrastructure, to reflect the evolving nature of the RFI threat. We consider it important to point to the reduction in redundancy, and therefore in safety, that stems from a systematic, permanent disruption of the primary means of navigation. Disruptions of GNSS are also known to increase controller workload and cause other forms of operational expenses, for example due to increased numbers of go-arounds.