DETECTION AND CHARACTERIZATION OF L1 BAND RFI USING SPACEBORNE GNSS-R SYSTEMS

M. Al-Khaldi^1,2, J. Johnson^1,2

¹The Ohio State University, ²ElectroScience Laboratory

Studies of terrestrial radiofrequency interference (RFI) sources within bands utilized for conducting microwave remote sensing of the Earth’s surface have had a long history of investigation. For example, in spite of the 1400 to 1427 MHz frequency being reserved for “passive space research”, “passive satellite Earth exploration” and “radio astronomy” the presence of RFI sources and their impact on the Aquarius, SMOS and SMAP missions has prompted numerous studies relating to detection, characterization, mitigation and geolocation of related transmitters over a multi-decade period. Conversely, studies of the RFI environment for more nascent Global Navigation Satellite System Reflectometry (GNSS-R) systems are less mature, motivating further analysis.

GNSS-R systems operate as a passive bistatic radars in which measured L-band forward scatter is cross correlated with local ‘clean’ replicas of navigation signals to form the fundamental GNSS-R measurement, the delay-Doppler Map (DDM). The DDM partitions received power from the spatial domain to an equivalent delay-Doppler space. An example is NASA’s 2016 CYGNSS (Cyclone Global Navigation Satellite System) Earth Venture Mission that includes eight GNSS-R observatories. GNSS-R systems like CYGNSS operate passively in the L1 band (1559 to 1610 MHz), reserved for transmissions by “radionavigation satellite systems” such that interference there can have implications that go beyond corrupting the quality of measured reflections.

This presentation explores a means by which interference signals within the L1 band can be detected using spaceborne GNSS-R observatories. The characteristics of the noise environment GNSS-R systems operate in are initially overviewed using an example commercial dataset as well CYGNSS’s record to highlight its various complexities. The utility of CYGNSS’s special raw I/F (Intermediate Frequency) mode for RFI detection is then demonstrated using the mission’s complete catalog of more than 510 acquisitions. The raw I/F data provides CYGNSS intermediate frequency data prior to correlation with the GNSS code or integration. The utility of kurtosis and cross frequency detectors is demonstrated with both suggesting that up to 25% of acquisitions were exposed to terrestrial RFI sources across all regions within the constellation’s coverage. Detections of interference sources with narrow and wideband spectral signatures and with low and high amplitudes are all observed. The results highlight how, in addition to the swath of disparate applications GNSS-R observations have been used for, some of their existing acquisition modes can be instrumental in better characterizing the GNSS RFI environment and aid in improved understanding of their sources in this vital, protected frequency band. This will also be complemented with a discussion relating to the geolocation of the sources of interference over the mission’s period of operation. It will be shown that up to 50% of the detections occur in the MENA (Middle East and North Africa) region as well as Myanmar. More intermittently operated sources are detected in Niger, Sudan, Venezuela, Guyana, Brazil and Argentina.