This is the distribution of numbers of circles intersecting in each cell. Very left-heavy.
Input on feasible stat-testing welcome.
Fortunately, the Baltic Jammer has quited down drastically now.
Last 7 days have only seen activity very limited but in coverage and in length of time.
A few beams against Gdansk airport, but that seems to be it. Still too few observations to say anything certain.
More persons are recording it now. A few people are working on other localization project. More to come unless Russian EW forces shut it down fully.
Many thanks to @lemonodor @rundradion @giammaiot2 @VelvetBlade @Dmojavensis @EyeofStolas for input.
The Baltic Jammer is a Russian GPS jammer that since Dec -23 has affected the navigation of 1000's of civilian & military aircraft.
Old method, new dataset, even more obvious.
Finally, data to disprove me. Thread 🧵
These maps depict density of radio horizons calculated from the first positions where 11496 aircrafts lost navigation from Feb to now. Altitude based method.
Below more zoom levels. There is no longer any doubt in my mind. The Jammer is in Kaliningrad, I believe at the coast.
I made this analysis twice before, with less refiend data.
Here is todays refined set but run with the same settings as in January. The Baltic Jammer is clearly in Kaliningrad.
Baltic Jammer has been running for 47 consecutive hours making this the longest run ever.
Red dots show affected area. Yesterday at 22:00Z it changed into a new configuration, green dots.
1614 unique airplanes affected, most of them civilian airliners.
Green dot map show the new configuration fully focused on almost exlusively Polish controlled airspace (Warzawa FIR), leaving a nice clear corridor for Kaliningrad-flights.
Here you can see the full metrics for last 3 days. These might need adjustment since the jammer is changing.
Today the Baltic Jammer changed. Starting with normal characteristic yesterday, after running for ~11h 0720Z there was a sudden change in area affected.
A NW sector of about 140°-180° is spared from GPS-jamming since then. It has made the problem almost exclusively Polish.
Still images of the periods before and after 0720 this morning. Each dot is a message of NIC 1-6 (low GPS quality).
The Estonia area is another jamming process, with other temporality/spatiality. It is almost always on.
Me and @rundradion discovered it about the same time from our plots.
Suddenly the metrics were different, not clear, not fully on.
When its off the % is at 0 with very few blips, see the period preceeding yesterdays start up.
To detect aircrafts having their GPS/GNSS jammed I have used the accuracy value NIC, and as a cutoff chosen 1-6 for individual flights. This has worked well and managed to find confirmed jammed flights.
But, the last days I have found suspiciously widespread results with this.
1-6 NIC reports to the border of my coverage: Italy, Switzerland, France....
This seems impossible. Indicating too high sensitivity.
I ran the numbers. Plot shows 'Blue: % NIC 1-6' and 'Green: Average NIC' Pink: new cutoff
Plot hints that my cutoff might be too un-sensitive!