Something is jamming GPS over Europe. Here's what we found

Scientific concepts / discoveries / nature phenomena

1) Continent-scale GNSS (GPS) jamming signatures

• Researchers analyzed a decades/years-spanning dataset from a network of GPS/GNSS monitoring stations that continuously records satellite navigation signal strength relative to background noise (signal-to-noise ratio, SNR).
• They found sudden, simultaneous drops in SNR across many receivers at the same exact moments, with:
  • ~10× reduction in signal-to-noise ratio during bursts
  • 75 days of similar disruptions since 2019
  • Events lasting only 3–5 seconds
• The disturbances showed Europe-wide reach, extending to:
  • Svalbard (north)
  • Spain (south)
  • eastern Poland / Kaliningrad region (apparent “center” pattern)
  • Canada (west)

2) Why the source was inferred to be “from space,” not on the ground

• Local interference (e.g., from towers/aircraft) was rejected because the effect was continental scale.
• By mapping which stations were affected and applying Earth-geometry constraints, the researchers inferred a high-altitude source:
  • Using conservative horizon/geometry assumptions, the jammer had to be at least ~1,200 km up (above the ISS altitude).

3) Solar storm comparison, then rejection of “natural solar cause”

• Solar activity is a known source of GNSS disruption:
  • A major solar storm in Nov 2025 disrupted GPS globally for hours.
• However, the observed events differed from typical solar behavior:
  • Solar interference generally builds/fades over tens to hundreds of seconds, but these were abrupt 3–5 s bursts
  • The disrupted band was narrow:
    • centered at 1,577.5 MHz
    • only about 5 MHz wide
  • Solar radio bursts affect the entire sunlit side of Earth, but these events were Europe-concentrated
• Conclusion: the source was most consistent with a satellite rather than the Sun.

4) GNSS / GPS positioning fundamentals (methodology described)

GNSS relies on timing and geometry:

• A receiver listens to GNSS signals containing:
  • satellite position in space
  • exact transmission time
• The receiver computes distance using signal travel time × speed of light.
• With satellites:
  • One satellite → position lies on a sphere
  • Two satellites → position reduces to a circle
  • Three satellites → reduces to two candidate points (only one is on Earth)
• With real-world receiver timing error:
  • phones must estimate receiver clock bias
  • thus, typically requires four satellites (four equations/four unknowns), and in practice more for error mitigation
• Accuracy improvements were attributed to:
  • relativity corrections
  • Earth rotation during signal travel
  • ionosphere and troposphere delay corrections
  • use of many satellites and continuous monitoring
• Ground station timing/positioning:
  • tied to an Earth reference frame built using quasars
  • quasar radio waves arrive at different telescopes with tiny timing differences
  • used to precisely map telescope (and thus station) relative positions and Earth motion

5) Jamming mechanism: why GNSS signals can be overwhelmed

• GNSS signals are extremely weak at Earth:
  • tens of 50 watts transmitted from high orbit
  • arriving at receiver at roughly ~10⁻¹⁶ W scale
• Jamming works by broadcasting a stronger signal in the same frequency band so navigation signals become indistinguishable from noise.
• The “sweet spot” referenced for GNSS relevance:
  • roughly 1.1–1.6 GHz (driven by atmospheric transmission and antenna considerations)
• Deliberately interfering is illegal in most countries.

---

Investigation methodology (as outlined)

• Step A: Retrospective anomaly search

  • Use public, older GNSS monitoring data (2019–2021 era dataset)
  • Check specific times for simultaneous SNR drops across the monitoring network
  • Identify repeated events (e.g., 75 disrupted days since 2019)

• Step B: Spatial consistency / geometry filter

  • Map affected receiver stations
  • Apply Earth curvature / line-of-sight constraints:
    • ground sources can’t cover all stations simultaneously
    • high-altitude source inferred (≥ ~1,200 km)

• Step C: Candidate satellite narrowing (single-satellite assumption approach)

  • Require a suspected satellite be above the horizon for all affected stations at once
  • Start with >15,000 active satellites
  • Eliminate >98% via constraints
  • Reduce to ~200 candidates
  • Further reduce using repeated alignment on multiple days → 14 suspects
  • Check public documentation and frequency compatibility for suspects

• Step D: Real breakthrough by higher temporal resolution (“raw voltage” sampling)

  • Standard receivers output simplified measurements once per second; jamming lasts only 3–5 s, so coarse data can’t compare arrival timing between stations
  • Researchers used specialized receivers to capture raw antenna voltage at tens of megahertz
  • Use two stations (Amsterdam, Netherlands; Trondheim, Northern Norway) to measure:
    • time difference of arrival (TDOA) between the same burst at two locations

• Step E: Hyperboloid localization

  • Convert TDOA into a 3D geometric locus:
    • a hyperboloid of possible jammer positions
  • High sampling rate yields small timing uncertainty → localization thickness about ~5 meters (in the geometric sense described)

• Step F: Exhaustive satellite testing against TDOA curve

  • For each satellite, use known orbit to predict expected TDOA
  • Eliminate satellites whose predicted timing difference doesn’t match observations
  • The test is “incredibly strict” because alignment must remain consistent as the source moves

---

Resulting identification of the interference source

• The analysis pointed to one satellite:
  • Cosmos 2546 (Russian satellite)
  • Matched the hyperboloid timing constraint within ~200 m (within uncertainties of public orbit data)
• Limitation acknowledged:
  • Cosmos 2546 launched May 22, 2020, so it can’t explain all events back to 2019
• Resolution:
  • Cosmos 2546 is part of a six-satellite constellation tied to Russia’s early missile warning system (described as part of “Golden Dome”)
  • These use Molniya orbits (highly elliptical), which:
    • linger over high latitudes
    • enable broader global coverage

---

Additional frequency clue suggesting multi-constellation / multi-system behavior

• A second interference burst was detected at:
  • 1,558.5 MHz
  • overlapping with the Chinese BeiDou navigation band
• This raised suspicion of testing or exercising capabilities relevant across systems.

---

Interpretation debate: intentional testing vs other short transmissions

Two competing interpretations were presented:

• Intentional test / capability exercise

  • The narrowband offset and periodic-like behavior suggest a deliberate operational “exercise” rather than randomness.

• Alternative hypothesis: very brief communications

  • Another team suggested the signals might be short communication messages from the satellites.
  • Jamming them could be strategically “protected,” since disrupting them could also risk interfering with one’s own navigation.

The subtitles do not claim definitive proof of intent—only strong inference from observed patterns.

---

Broader implications and mitigation concepts

• Potential impact if fully deployed

  • GNSS supports aviation, shipping, network timing, logistics, and more
  • A space-based jammer could disrupt regions too large to shield easily at ground level

• Other GNSS interference type mentioned: spoofing

  • GPS spoofing swaps/imitates signals to feed a plausible but false location
  • Reported as affecting >1,500 flights per day in the subtitles

• Resilience strategy: multi-source PNT architecture

  • Suggested resilient architecture includes:
    • space-based signals
    • terrestrial broadcast
    • fiber timing
  • Mentioned alternatives:
    • eLoran (high-powered terrestrial radio navigation)
    • possible navigation using magnetic/quantum systems (Earth magnetic field variations)

---

Researchers / sources featured (explicitly mentioned)

• Todd Humphreys — Professor, GPS/GNSS expert, University of Texas at Austin
• Zach Clements — student, University of Texas at Austin (worked with Humphreys)
• Gregor — appears in dialogue; full name not provided
• Ramsey — appears in dialogue; full name not provided
• Derek — appears in dialogue; full name not provided
• German Aerospace Center (DLR / DLR-related team) — proposed real-time tracking and interception effort
• Independent teams in Europe — reported verification of aspects (no specific names given)
• Ground News — sponsor (media monitoring/data service; not a research source for the GNSS findings)

---

Specific satellite / system names mentioned

• Cosmos 2546 (Russia)
• Russian missile warning constellation described as part of “Golden Dome”
• Molniya orbits (orbital type used by the constellation)
• BeiDou (China GNSS)
• GPS (US GNSS)
• Other GNSS mentions:
  • Russian navigation system
  • European navigation system (Galileo)
  • Chinese navigation system (BeiDou)