Investigation of Potentially Critical Interference Environments for GPS/Galileo Mass Market Receivers

Kurzfassung

Positioning and timing services provided by the U.S. Global Positioning System (GPS) have already given rise for various applications almost in all fields of our everyday life. With the advent of European Galileo significant performance improvements for civil users are expected due to the doubled number of satellites and the availability of multiple open signals. The L1/E1 frequency band is expected to be the most probable candidate for mass market GPS/Galileo receivers. It is currently a common belief that the L1/E1 band is relatively quiet in terms of radio frequency interference. However, evidence or counterevidence of this belief is still an open issue. The opportunity to prove this belief has occurred during an interference measurement campaign carried out by the German Aerospace Center (DLR) within the framework of the GJU project GIRASOLE [1]. For this purpose a dedicated measurement van built and further adapted to DLR’s needs by Joanneum Research in Austria was provided by ESA. It was equipped with one Rohde Schwarz FSH6 and one Agilent E4443A spectrum analyzers, several directional and one hemispherical antenna and one dedicated measurement unit by Joanneum Research. The measurements have been carried out in exemplary large cities in Germany (Munich, Augsburg, Hamburg) and were primarily focused on investigating the radio interference situations in the new Galileo frequency bands corresponding to the safety of life services (E5/E1). The L1 band among others has been systematically scanned for the presence of radio interferers. As an interesting result of this campaign, it turned out that in several urban area scenarios high power pulsed interferers were partially situated in L1. The interference signals observed have bandwidths of several MHz and were detected at different frequency offsets from L1 centre frequency. Roughly speaking, these interferers are invisible for non-professional GPS L1 receivers with low RF-Front -End bandwidths but may cause problems to future GPS/Galileo receivers designed to operate with broader bandwidth signals. For example the interference signals were observed to partly overlap with the frequency range of 4 MHz around 1575.42 MHz which contains most of energy of future Galileo E1B and E1C BOC(1,1) Open Service signals. This paper aims at assessing the sensitivity of GPS and Galileo receivers with respect to pulsed interferers observed near L1 band. Two different approaches will be applied for the assessment using both software and hardware simulations. Following the first approach, we use GPS/Galileo acquisition and tracking MATLAB software for determining the corresponding levels of maximum tolerable interference under interference conditions similar to those observed within the GIRASOLE measurement campaign. When performing simulations we assume that the receivers use RF-Front-End configurations (e.g. 2 or 4 bit ADC, 4 MHz bandwidth etc.) which are typical of mass-market GPS/Galileo receivers. Moreover, since mass-market GPS/Galileo receivers are considered, standard signal processing algorithms are applied for acquisition and tracking without utilizing dedicated high sophisticated methods for radio interference mitigation. To enable realistic simulations of pulsed interference signals a parametric interference signal model is developed based on the results of the measurement campaign using characterization both in time and frequency domain. The parametric model easily allows for studying the sensitivity of GPS/Galileo signal processing to different parameters of the interfering signals like pulse repetition rate, power and spectrum separation from the L1 centre frequency. Following the second assessment approach, hardware simulations are performed making use of the combination of the modified Spirent GSS 7790 GNSS Simulator and the Agilent E8267D Signal Generator. The digital signal samples containing the detected L1 interferers are used to create RF L1 interference signals which act as external interference source in the Spirent GNSS Simulator. The output of the simulator, which is a mixture of radio interference signals and signals of GPS satellites, is fed into a hardware GPS receiver in order to determine the effect of the interference signals onto the acquisition and tracking performance and to identify critical interference levels. The results of the hardware and software simulations are compared. The work tends to provide valuable insights in and results about the degree of interference threat related to pulsed interference signals partially overlapping with L1 GPS/Galileo. The assessment of maximum tolerable levels of this interference for mass-market receivers without dedicated mitigation algorithms will be provided. Extensive software and hardware simulations are carried out to corroborate the gained results.