[ad_1]
Bergman had previously spotted fake tracks on the AIS website when virtual sailboats from an online sailing racing game probably appeared on AISHub last year. But it was the first time he had seen real ships and the warships were no less so.
“At SkyTruth we are particularly concerned when false data affects fishing,” Bergman said in an interview in a video call. “But overall we want to understand how data is falsified and what we can do to detect and correct it.”
Bergman identified nine warships from a screen in the story, and then compared the fake AIS messages with the actual messages sent by the same ships before and after the scammers. It was immediately noticed that there were no amateur bumps or accidents. “The false messages were very plausible because we confirmed that the Swedish navy’s position was false,” he says.
There are more than 20 types of AIS messages — some for supertankers, some for ships — and each has a wide range of data areas, from navigation information to unknown communication settings. Comparing areas that are usually invisible to sailors are narrow, Bergman eventually found subtle differences between false and true data. He then used this model to write a query for the global historical database of AIS messages — and was impressed with the results.
His search found nearly a hundred sets of messages from multiple AIS data providers, ranging from last September to thousands of miles. More worryingly, the damaged ships were almost exclusively military ships from European and NATO countries, including at least two U.S. submarines.
“It was worrisome when I also noticed a lot of other vessels that showed this unusual AIS profile,” says Bergman. But he should have known that the suspicious AIS messages were actually fake, not the result of a technical excitement or a special military deployment. Bergman spent the next few months hardly checking the actual positions of the ships they were destined for. Initially, he used open source data such as news, military press releases, and fan websites like Warshipcam.com. “A lot of people like to take pictures of seagoing ships and post them online,” Bergman says. “I found examples of ships leaving or entering areas that seemed quite impossible.”
Bergman covered the synthetic aperture radar and optical images of the European Space Agency Sentinel-1 and –2 satellites to suspicious AIS pings. If they were true, the AIS data would have to match the satellite image of the vessel.
Instead, Bergman saw only the empty ocean, over and over again. In fact, he says, “I still haven’t found an instance when a clue that the query falsified was real.”
The stars didn’t always line up for Bergman’s detective work. Some AIS tracks do not match the top of the ESA satellite, or the optical images fell on cloudy days when they were useless. And some warships didn’t have a lot of Instagram fans. In the end, Bergman managed to confirm that about 15 AIS data sets are definitely fake.
As well Queen ElizabethDue to H’s imaginary flotilla, Bergman found false traces of other warships from the US, the Netherlands, Belgium, Germany, Lithuania, Estonia and Sweden. A previously unreleased suspicious runway shows the USS being the destroyer of US guided missiles Roosevelt by steaming 4 kilometers of water around Russian territory around Kaliningrad last November, if the maneuver was genuine it would be prudently provocative. It appears that there were five other fake attacks around Kaliningrad in June. One was a Polish warship that followed the same track, speed and route as a Swedish corvette five days earlier, another way for Bergman to say that the tracks were digitally created.
[ad_2]
Source link
