Frequency Domain Adaptive Filtering Against Pulsed Interference: Performance Analysis Over Europe

Civil Aviation standardisation bodies (ICAO, RTCA, EUROCAE) are currently investigating the use of Global Navigation Satellite Systems (GNSS) as a stand-alone navigation solution for civil aircraft. For obvious safety reasons, on-board GNSS receivers must guarantee minimum requirements in given phases of flights. These requirements, dependent upon the system and signals used, are stated in the Minimum Operational Performance Specification (MOPS), published (or being published) by the corresponding authorities. With that respect, the future use of Galileo E5 and GPS L5 bands has raised, among others, interference issues. Indeed, pre-existent RF systems emit in this band, though interfering with the E5/L5 signals. The main threat was identified as being DME/TACAN ground beacons pulsed emissions [RTCA, 2004]. Without any mitigation capability, these systems can disturb the proper functioning of on-board GNSS receivers, preventing them from complying with safety requirements. Two Interference Mitigation Techniques (IMT) have been proposed to fight this threat, the Temporal Blanker (TB) and the Frequency Domain Adaptive Filtering (FDAF). The TB technique [Grabowsky, 2002] offers a fairly simple implementation and was shown [Bastide, 2004] to provide enough benefits to ensure that the specified requirements were met in all phases of flights for a GPS L5 or Galileo E5 receiver. However, it was also demonstrated that the resulting performances were meeting the requirements by only a small margin on the worst locations. In contrast, the FDAF is a more demanding mitigation technique against pulsed interference in terms of required resources but [Raimondi, 2006] showed that it could bring a stronger margin with respect to the civil aviation requirements. Simulations [Raimondi, 2006] showed that the use of FDAF allowed an decrease of the postcorrelation C/N0 over the worst DME/TACAN interference environment that can be found in Europe, so called the European "hot spot". However, simulations obtained through running an accurate model of a GNSS receiver are often extremely time-consuming, and, in this condition, it is a cumbersome work to draw a world map checking the good functioning of FDAF in all locations. It is then interesting to find a simplified theoretical way to derive the post-correlation C/N0 degradation suffered by the signal as a function of the interference environment, using an FDAF technique. The aim of this article is to describe a new simulation tool aiming at calculating this degradation in all location in a very short time. Results are first confirmed and then compared to temporal blanker's one.