GNSS jamming zone mapping from space
Deliberate GNSS jamming raises the L-band noise floor across wide areas. Spaceborne RF receivers can detect, geolocate and map those elevated noise zones in near-real-time, giving governments and operators a picture that no ground network alone can provide.
Sensors
- Spire Global GNSS-RO / GNSS-reflectometry constellation: Over 100 LEO CubeSats carrying geodetic-grade GNSS receivers. Originally designed for radio occultation, they also record raw signal-to-noise ratio and carrier-phase data that reveal L1/L2 noise floor elevations consistent with jamming. Orbital geometry gives multiple passes per day over most mid-latitude regions, though exact revisit at a single point depends on constellation density and inclination mix.
- HawkEye 360 RF-monitoring constellation: Clusters of three formation-flying satellites carrying wideband RF receivers covering roughly 144 MHz to 15 GHz, which spans GNSS L-band (1.176–1.602 GHz). Time-difference-of-arrival (TDOA) and frequency-difference-of-arrival (FDOA) between cluster members geolocate emitters to a published accuracy of around 1 km CEP under favourable geometry, with revisit at any given point typically measured in hours rather than minutes.
- GAL-EGNOS RIMS ground network (reference / ground truth): The European Geostationary Navigation Overlay Service Ranging and Integrity Monitoring Stations form a fixed continental network. They cannot geolocate jammers independently, but their logged carrier-to-noise-density (C/N0) drops provide time-stamped ground truth against which spaceborne detections are validated.
- Galileo and GPS satellite signal health broadcasts: Both constellations broadcast integrity messages and signal anomaly flags. Aggregated receiver autonomous integrity monitoring (RAIM) fault reports from aviation and maritime users, when collected systematically, create a crowdsourced noise-floor map that complements dedicated RF monitoring payloads.
What a jammer actually does to the L-band environment
A GNSS jammer works by flooding the receiver's front end with broadband or swept noise centred on one or more of the civil GNSS frequencies: GPS L1 at 1575.42 MHz, GPS L2 at 1227.60 MHz, and Galileo E1/E5 across 1176–1575 MHz. The jammer does not need to be powerful. Because genuine GNSS signals arrive at roughly minus 130 dBm after travelling 20,000 km from medium Earth orbit, even a cheap 0.5-watt jammer can suppress reception across a radius of several kilometres at ground level.
From orbit, the geometry inverts. A spaceborne receiver at 500 km altitude looking down at a jammer sees the signal growing stronger as the satellite approaches, then fading as it recedes. That characteristic rise-and-fall in the L-band noise floor, timed against the satellite's precisely known ephemeris, is the primary detection signature. It is distinct from natural ionospheric scintillation, which is covered separately in this library's page on ionospheric monitoring.
How space-based receivers localise the source
Formation-flying satellite clusters, such as those operated by HawkEye 360, measure the time at which the same jammer pulse arrives at each spacecraft. Because the satellites are separated by tens of kilometres in their cluster, the arrival-time differences, combined with Doppler shift differences (FDOA), constrain the emitter's ground position. With a three-satellite cluster and good geometry, published figures suggest geolocation accuracy of roughly 1 km CEP, degrading to several kilometres when the emitter is near the satellite ground-track edge or when the jammer duty cycle is intermittent.
Single-satellite approaches use a different technique. By recording the received power profile across a full overpass and correlating it against the satellite's velocity and altitude, analysts can estimate the jammer's range and bearing, though without a second independent line of bearing the result is an arc rather than a point. Multiple overpasses from different orbital planes, or data fusion with ground-network C/N0 anomalies, resolve the ambiguity. Neither method is instantaneous: a HawkEye 360 cluster overpass lasts a few minutes, and latency from collection to processed geolocation has historically been measured in hours, not seconds.
What the Eastern Mediterranean and Baltic cases have shown
The Eastern Mediterranean has become the most documented GNSS jamming environment outside a controlled test range. Since at least 2019, aviation safety bodies and open-source researchers have logged persistent jamming affecting aircraft approaching Cyprus, Lebanon and Israel, with affected zones extending hundreds of kilometres from apparent source areas. The Nordic countries, particularly Finland and Norway near the Kola Peninsula, have recorded episodic jamming events correlated with military exercises, with Norwegian aviation authority AVINOR publishing formal notices to airmen (NOTAMs) on several occasions.
These events share a pattern: jamming tends to be geographically stable over days or weeks when associated with a fixed installation, but highly mobile when associated with vehicle-mounted or vessel-mounted systems. Spaceborne RF monitoring is particularly useful for the mobile case, because a ground reference station only sees what passes through its antenna beam, while a LEO satellite sweeps the entire affected region. The Baltic Sea environment has also demonstrated that maritime vessels can carry jammers, creating moving interference zones that confound port-approach navigation and AIS position reporting simultaneously.
Honest limits of the method
Revisit is the principal constraint. Even a large commercial RF-monitoring constellation cannot provide continuous coverage of a single location. A point of interest in the Eastern Mediterranean might receive two to four useful cluster overpasses per day from HawkEye 360, meaning a jammer that operates for only 30 minutes between passes is invisible to the spaceborne layer. Ground networks fill this gap for fixed installations, but most governments do not operate dense enough RIMS-equivalent networks in contested regions.
Geolocation accuracy degrades with jammer behaviour. A jammer that sweeps frequency rapidly, or that uses a very low duty cycle, produces a weaker and noisier TDOA signature. Low-power personal privacy devices, sold illegally in many countries, may fall below the detection floor of current spaceborne receivers entirely; published sensitivity figures for HawkEye 360 are not public, so the minimum detectable effective radiated power remains an honest unknown for open-source users.
Attribution is a separate problem from detection. Knowing that a jammer is operating at a given grid reference does not identify who controls it. That inference requires corroboration from other intelligence layers.
Turning detections into a usable intelligence product
Raw geolocation fixes become operationally useful only after aggregation, filtering and contextualisation. A single fix on a given day might be noise or a misidentified civil emitter. A cluster of fixes at the same grid reference across ten consecutive days, with consistent power levels and frequency characteristics, is a credible persistent jammer. The analytic workflow therefore involves: ingesting raw RF detection events, clustering them spatially and temporally, computing confidence scores based on fix count and geometry quality, and overlaying the result against known infrastructure, vessel tracks and aviation corridors.
Governments commissioning this kind of monitoring typically want two outputs: a near-real-time alert when a new jamming zone appears, and a periodic map showing zone boundaries, estimated effective radiated power (where derivable), and trend over time. Satellize structures similar spatial-analytics pipelines for clients requiring sovereign data custody, drawing on the same data-fusion principles used in its Tonga crop-estimation programme, where multi-source inputs are reconciled into a single calibrated output layer. The delivery format matters as much as the analysis: aviation authorities generally want NOTAMs-compatible zone polygons, while defence clients typically require GeoJSON or shapefile layers ingested into their own command systems.
What a procurement team should ask before commissioning
Three questions cut through most of the ambiguity in this market. First, what is the actual revisit rate over your specific area of interest, not the global average? Constellation operators quote impressive aggregate statistics that may not apply to a narrow latitude band or a small island. Ask for simulated pass schedules over your AOI before signing.
Second, what is the geolocation accuracy under realistic jammer behaviour, including intermittent and frequency-agile emitters? Published CEP figures for cooperative or continuous emitters are optimistic baselines. Third, what is the data latency from collection to delivered product, and does that latency meet your operational decision cycle? A 12-hour latency is acceptable for strategic trend mapping; it is not acceptable for real-time airspace management. Getting honest answers to these three questions will do more to scope a programme correctly than any amount of marketing material.
Typical figures
| Geolocation accuracy (TDOA/FDOA, cluster method) | ~1 km CEP under favourable geometry; degrades to several km at poor viewing angles or with intermittent emitters |
| Frequency coverage | L-band GNSS: GPS L1 (1575.42 MHz), L2 (1227.60 MHz), Galileo E1/E5 (1176–1575 MHz); HawkEye 360 wideband payloads cover ~144 MHz to 15 GHz |
| Revisit at a single point (HawkEye 360) | Typically 2–4 cluster overpasses per day at mid-latitudes; varies with constellation growth and orbital geometry |
| Detection latency (collection to product) | Hours (typical commercial pipeline); near-real-time tasking possible with direct downlink arrangements |
| Minimum detectable ERP | Not publicly specified by HawkEye 360; low-power personal privacy devices (<0.1 W ERP) may fall below detection threshold |
| Spatial extent of mapped zone | Single jammer ground-level exclusion radius typically 2–50 km depending on ERP; spaceborne swath covers hundreds of km per pass |
| Archive depth | HawkEye 360 commercial archive from ~2019; Spire GNSS data from ~2016; EGNOS RIMS logs vary by national authority |
| Delivery formats | GeoJSON, shapefile, KML zone polygons; CSV event logs; NOTAM-compatible zone descriptions on request |
| Ground-truth reference | EGNOS RIMS C/N0 logs; aggregated RAIM fault reports from aviation and maritime AIS receivers |
Analytics Satellize can run
| Persistent jammer zone map | Spatial and temporal clustering of TDOA/FDOA geolocation fixes; confidence scoring by fix count and geometry quality | Weekly GeoJSON polygon layer with confidence score, estimated ERP range and date-of-first-detection per zone |
| New-zone alert | Anomaly detection against rolling baseline noise-floor model; threshold trigger on first confirmed cluster of fixes at a previously clean grid cell | Near-real-time alert message with grid reference, frequency band, detection time and preliminary confidence score |
| Mobile jammer track reconstruction | Sequential geolocation fixes linked by Kalman-filter track association; corroborated against AIS vessel positions where maritime origin is suspected | Track shapefile with timestamped position estimates and associated vessel candidate list |
| Aviation corridor risk assessment | Intersection of jammer zone polygons with published ICAO airways and approach procedures; severity scored by zone area, C/N0 degradation depth and proximity to threshold altitude | Corridor risk report in NOTAM-compatible format; updated on each new detection cycle |
| Trend analysis and campaign attribution | Time-series aggregation of zone activity; change-point detection to identify start/end of jamming campaigns; overlay against known exercise calendars and open-source event data | Monthly intelligence summary PDF with annotated timeline, zone evolution maps and assessed campaign boundaries |
| Ground-network gap analysis | Modelling of existing RIMS or equivalent receiver coverage against detected jammer locations; identification of ground-station placement that would close coverage gaps | Coverage gap map with recommended receiver siting coordinates and expected improvement in detection latency |
Who does the work
We can get this done for you. Satellize runs its own analyst desk and a strong science team. You do not buy a data feed and work out what it means; our people source the imagery, run the analysis described on this page, and hand you the answer with its confidence limits stated. Discuss this requirement.