An Interference Monitoring System for GNSS Reference Stations

J. Wendel, C. Kurzhals, M. Houdek, J. Samson

Abstract:	Astrium GmbH and Iguassu Software Systems a.s. have developed an Interference Monitoring System (IMS) for the European Space Agency (ESA), which is designed to protect Galileo (GSS) and EGNOS (RIMS) sensor stations by providing situational awareness on the interference environment. This is of paramount importance for the operation of GNSS systems: Undetected degradations in sensor station measurements potentially impact the quality of the information provided via the navigation / service messages, which in turn can affect the whole user community. The IMS consists of a Processing Facility (PF) and several Local Elements (LE). The LEs act as sensor nodes which autonomously monitor the local interference environment and analyze interference events. The PF is the “central node” of the system, it collects and stores the information provided by the LEs and allows to instantly view the interference environment at every LE. Furthermore, the PF allows to configure the LEs from remote according to the user’s needs, so a local operator at the LEs is not required. Each LE consists of an outdoor unit including an antenna and appropriate signal conditioning equipment, and an indoor rack. The indoor rack contains a RF frontend which is connected to the outdoor unit and provides the RF signal to a signal analyzer. A NTP server allows for precise absolute time tagging of the signal analyzer measurements, and a remote power switch allows to power up and down every component in the rack, also remotely from PF. A rack-mount server is used to analyze the spectrum and the IQ data measurements provided by the signal analyzer. Finally, a rack-mount console (LCD, keyboard and touchpad) is included to allow for operation and maintenance of the system by a local operator, too. Currently, three LEs are deployed, one at the EGNOS RIMS in Warsaw, one at the GCC in Oberpfaffenhofen, and one at ESA ESTEC in Noordwijk. The PF is located as ESA ESTEC, too. This IMS prototype was operated for a six-month demonstration period to prove the functionality of the system, which was completed by the end of July 2012. Since then, the system has taken up normal operations. Each IMS LE monitors the frequency range from 900 MHz to 1800 MHz, but could be used to monitor much higher frequency bands, too, given that the bandpass filter in the outdoor unit is adapted accordingly. The principle interference monitoring process consists of two steps: First, spectra measurements covering the complete frequency range of interest are acquired. Hereby, maximum and average detectors are used, as these have complementary characteristics regarding their detection capabilities: Pulsed interferences with a low duty cycle might not significantly increase the average power within the interference bandwidth, therefore a detection of these interferences using the average detector is not possible, while these interferences might be easily visible using the maximum detector. On the other hand, the maximum detector suffers from an increased noise floor, burying interferences of lower power, which might be visible using the average detector, of course depending on their power level and duty cycle. Different detection algorithms are used to search the acquired spectra for interferences, for example assessing variations of the spectra power levels within a certain bandwidth, or checking for the excess of an appropriate threshold. The result of this process is a list of frequency bands, where interferers are suspected. In the second step, IQ data is acquired in each of these frequency bands, which is then analysed to confirm the presence of interferences and to identify the interference characteristics. Hereby, the Welch periodogram is used to determine the number of spectrally separated interferences within the acquired IQ data bandwidth, and to identify center frequency and bandwidth of each of these interferences. Then, a bandpass filtering is performed in order to obtain IQ data with all interferences wiped out, except for the one of which the time domain characteristics shall be identified. In order to identify the time domain characteristics of the remaining interference, the instantaneous power is calculated, which reveals whether this interference is continuous or pulsed, and for the latter case also provides the time intervals during which a pulse is present. Depending on power level and pulse duration, the pulses are assigned to different pulse groups, each of which is considered to be a separate interferer. For each of these pulse groups, the duty cycle is determined, and, in case of regular pulse intervals, also the pulse repetition time. Furthermore, special routines are used to detect double pulses like DME/TACAN. Finally, the instantaneous frequency is analysed, which allows to identify different types of chirps. The identified interference characteristics are compared with a database in order to associate the observed interference with a specific RFI source, where possible. Another important feature of the IMS is the prediction of the impact of interferences on the performance of GNSS receivers. From the identified interference characteristics, the degradation in C/No, the code jitter, the carrier jitter, and the impact of the interference on the availability of the observables and the possibility for re-acquisition is calculated. These predictions are based on closed form analytical expressions, which take into account the specific receiver parameters like front-end bandwidth, early-late spacing, loop filter bandwidth, pre-detection integration time and pulse blanking threshold, as well as the GNSS signal code rate. These parameters can be configured for each GNSS band separately. The impact prediction results allow for a fast detection of potential problems at the Galileo (GSS) and EGNOS (RIMS) sensor stations where the IMS LEs are installed. All information collected by a LE is used to generate a detailed report on the local interference environment with a summary at a daily, monthly, and yearly basis. These reports include status information on the LE itself, statistical information like the 95%, 99% and 100% percentiles of the spectra power levels, and detailed information on the time and frequency domain characteristics of critical interference events, as well as their predicted impact on the reference station receiver. At the PF, reports are generated which provide a general overview of this information for all LEs. In this paper, the IMS architecture is described. The approach for interference detection, characterisation, and impact prediction on reference station receiver performance is presented in detail, too. Furthermore, selected results on real-life interference events observed during the demonstration phase are shown, which illustrate the capabilities of the IMS system.
Published in:	Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013) September 16 - 20, 2013 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	3391 - 3398
Cite this article:	Wendel, J., Kurzhals, C., Houdek, M., Samson, J., "An Interference Monitoring System for GNSS Reference Stations," Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013), Nashville, TN, September 2013, pp. 3391-3398.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In