A year-long project aboard a commercial cargo ship collected tens of thousands of snapshots of radio-frequency interference in the GNSS band on a passage from Spain to Korea and back. Most interference was detected in busy port areas, less interference while transiting along coasts, and while least frequent, interference was still found in the open ocean.

Researchers at the German Aerospace Center (DLR) are still analyzing the vast amount of GNSS disruption data collected during the year-long project. Two papers have already been published about this project, and more are on the way, according to principle researcher Emilio Pérez Marcos.

In a paper presented at the Institute of Navigation last year, Marcos and his co-authors outlined the results of the last five months of this unique sampling experiment. Detection equipment was mounted on a large Hapag-Lloyd container ship. The antenna was mounted about 50 meters above the water line and provided a line-of-sight of 25km or more. The L1/E1 and L5/E5a frequency bands were continuously monitored. In addition to a “Snapshot” recording device used to save raw data samples (time snapshots), a more resilient DLR multi-antenna receiver was used to assess the impact of interferences in beamforming array GNSS receivers (semi-resilient).

As might be expected, the most interference was detected in busy port areas. Less interference was experienced while transiting along coasts. While it was the least frequent, interference was still detected during open ocean transits.

Of the 39,045 snapshots recorded, 6,632 contained radio frequency interference at 1dB or higher. Separate tests have shown that many single antenna GNSS receivers begin to perform poorly with interference signals greater than 1dB. The other 32,413 snapshots could represent interference signals that may have come from weaker transmitters, sources more distant from the ship, been the result of adjacent band transmissions, or other phenomena.

Three particularly strong and persistent interference incidents were noted in the paper.

The first was detected when the vessel was transiting the Suez Canal northbound. The interference lasted around five hours and 60km. At several points the interference prevented the DLR semi-resilient GNSS receiver from working properly, which would mean that any single antenna GNSS receiver would cease to function completely.

The second caused the DLR receiver to fail when the vessel was entering Jebel Ali, the port of Dubai in the United Arab Emirates. The DLR receiver provided some resilience thanks to its beamforming capabilities; again any other receiver would have suffered the interference effects earlier being unable to provide any PVT. The receiver did not return to proper operation for 11 days and 5,000km. The reason for this is uncertain and under investigation.

Particularly strong interference (45dB) caused the third incident and resulted in the DLR receiver failing for three days. It began when the ship was entered the highly trafficked Malacca Straits.

The equipment used also allowed researchers to determine direction of arrival for the interfering signals and to evaluate whether the interference was a spoofing signal.

For the reported strong interference events, DLR consulted the captain of the ship, who attested and confirmed the loss of PVT in the ship’s own GNSS receiver, with all the consequences that this implies for the systems that rely on it.

The paper, “Interference and Spoofing Detection for GNSS Maritime Applications,” was presented at the ION GNSS+ conference in Miami in September of 2018. It described the last phase of a yearlong measurement effort aboard the ship by DLR. An earlier phase of the campaign has also been published in E. P. Marcos et al., “Interference awareness and characterization for GNSS maritime applications,” 2018 IEEE/ION Position, Location and Navigation Symposium (PLANS), Monterey, CA, 2018.

The authors are preparing additional papers to describe more of the results from the larger project.