L-band mobile satellite service congestion and occupancy monitoring
Spaceborne RF scanners can measure power spectral density across the 1–2 GHz L-band from LEO, exposing congestion, unauthorised access and interference events in mobile satellite service allocations that ground-based monitoring consistently misses.
Sensors
- HawkEye 360 cluster satellites: Three-satellite formation flying at roughly 575 km altitude, measuring time-difference and frequency-difference of arrival across 15 MHz–3 GHz. Geolocation accuracy published as approximately 1 km CEP for strong emitters; revisit over a given ocean cell varies from 30 minutes to several hours depending on orbital geometry. Covers L-band MSS allocations directly.
- Spire STRATOS RF payload: Software-defined radio payload hosted on Spire's LEMUR-2 bus, capable of scanning configurable frequency windows including the 1525–1660.5 MHz Inmarsat downlink and 1626.5–1660.5 MHz uplink bands. Spire operates over 100 satellites, giving sub-hourly revisit over most ocean areas, though dwell time per frequency window is limited by scanning duty cycle.
- Inmarsat published interference reports: Inmarsat files interference reports with the ITU and publishes operator advisories identifying uplink interference events on its L-band GEO fleet (I-4 and ALPHASAT satellites operating at 1626.5–1660.5 MHz). These reports provide ground-truth timestamps and approximate geographic regions for cross-validating LEO intercept data.
- Iridium network telemetry (published): Iridium's 66-satellite LEO constellation operates at 1616–1626.5 MHz. Iridium publishes network availability and interference event data through its maritime safety communications role. Its cross-linked architecture means an interference event on one satellite can propagate measurably across adjacent links, providing a secondary detection layer.
Why L-band is both essential and contested
The 1–2 GHz L-band is the workhorse of global maritime and aeronautical safety communications. Inmarsat's GMDSS services, Iridium's satellite phone network, and the distress and safety functions mandated under SOLAS all depend on allocations within this narrow slice of spectrum. The ITU Radio Regulations, specifically Article 9, govern coordination between these mobile satellite services and other primary users, but coordination on paper does not prevent real-world congestion or deliberate interference.
The practical problem is density. In the Strait of Malacca, the Gulf of Aden, or any major port approach, hundreds of vessels simultaneously attempt uplink access on shared L-band channels. Congestion degrades voice quality and, more seriously, can delay or corrupt distress messaging. Unauthorised transmitters, whether misconfigured terminals or deliberate jammers, are difficult to locate from the ground because the interfering signal arrives at the GEO satellite from below, mixed with legitimate traffic. A receiver in LEO, looking down at the same geometry, sees the emitter directly.
The physics of looking down instead of up
A GEO satellite receiving an uplink interference event sees it against the thermal noise floor of its own receiver, with the interfering signal having traversed roughly 35,786 km. A LEO scanner at 500–600 km altitude intercepts the same uplink signal at approximately 1/4,700th of that path length. The free-space path loss difference is around 73 dB. That is not a marginal advantage; it means a LEO receiver can detect emitters that are completely invisible to the GEO operator's own telemetry until they are strong enough to cause measurable degradation.
Power spectral density measurement from LEO therefore gives a pre-symptomatic view of congestion. A cluster of vessels all transmitting on the same Inmarsat uplink sub-band produces a characteristic spectral signature: elevated noise floor, reduced guard-band separation, occasional capture of individual carrier frequencies. HawkEye 360's formation geometry allows time-difference of arrival (TDOA) and frequency-difference of arrival (FDOA) geolocation, placing individual strong emitters to roughly 1 km CEP under good signal conditions. Weak or intermittent emitters are harder; the honest floor is closer to 5–10 km for low-power terminals.
What congestion looks like in the data, and what it does not
Occupancy monitoring produces a time-series of channel occupancy percentage across the monitored sub-bands. A well-managed L-band allocation in open ocean might show 20–40% occupancy during peak hours. The same allocation over a major anchorage can saturate above 80%, with individual carrier detections overlapping in frequency. The distinction between congestion and interference is important: congestion is a lawful condition that degrades performance; interference implies an emitter operating outside its authorised parameters, whether in frequency, power, or geographic zone.
Spaceborne monitoring cannot always resolve that distinction from spectral data alone. A carrier sitting 50 kHz outside its assigned slot could be a misconfigured terminal, a Doppler-shifted legitimate signal, or a deliberate intrusion. Attribution requires cross-referencing with vessel position data (AIS, where available), ITU filing databases, and, where the emitter is persistent, multi-pass TDOA geolocation to build a location history. Cloud cover is irrelevant to RF collection, which is one of the few genuine advantages this method has over optical surveillance. The main operational limit is revisit: a single HawkEye 360 overpass of a given ocean cell lasts a few minutes, and the gap before the next pass can exceed an hour in some orbital configurations.
Regulatory context and what operators actually need
Maritime safety regulators and MSS operators face a shared but differently weighted problem. The regulator, typically a national maritime authority or the ITU Radiocommunication Bureau, needs evidence of interference events sufficient to initiate coordination or enforcement under Article 9 procedures. That means timestamped, geolocated detections with spectral evidence, not just a vessel master's complaint. The MSS operator needs faster intelligence: which uplink beams are congested, where the interference is originating, and whether it is growing.
Spaceborne occupancy data can serve both. For the regulator, archived multi-pass intercept data provides a defensible evidentiary record. For the operator, near-real-time occupancy feeds allow dynamic beam management and early warning of emerging interference sources. Neither use case requires the LEO scanner to decode the content of the transmissions, which avoids the legal complications of interception law in most jurisdictions. Power, frequency, and location are sufficient.
Honest limits: what this method cannot do
Geolocation accuracy degrades significantly for low-power, short-duration transmissions. A single-burst Iridium call lasting a few seconds may not provide enough signal for reliable TDOA across a three-satellite cluster. Spire's scanning architecture means any given frequency window receives dwell time measured in seconds per pass, which can miss transient events entirely. Neither system currently publishes detection sensitivity figures for L-band specifically in open literature, so minimum detectable signal estimates should be treated as indicative rather than guaranteed.
There is also a spectrum coordination gap: the ITU filing database records authorised allocations, but it is not a real-time operational picture. Cross-referencing intercept data with filings identifies candidates for unauthorised operation, but confirmation requires additional steps. Satellize incorporates this workflow into its analytics pipeline, where relevant, alongside the occupancy feeds it can run from commercial RF constellation data. The Tonga crop-estimation programme is a different domain entirely, but the underlying principle of turning raw satellite data into a decision-ready output is the same.
Typical figures
| Frequency coverage | 1525–1660.5 MHz (Inmarsat L-band uplink and downlink); 1616–1626.5 MHz (Iridium); broader 15 MHz–3 GHz scan range available on HawkEye 360 |
| Geolocation accuracy (strong emitters) | Approximately 1 km CEP (HawkEye 360, published); 5–10 km for low-power or intermittent terminals |
| Revisit over a fixed ocean cell | 30 minutes to several hours depending on constellation geometry and latitude; Spire's 100+ satellite fleet improves frequency but dwell per pass remains short |
| Detection method | Power spectral density measurement; TDOA/FDOA geolocation from formation-flying or multi-satellite intercept |
| Minimum detectable signal | Not published for L-band specifically; LEO path-loss advantage over GEO is approximately 73 dB, enabling detection of emitters below GEO noise floor |
| Latency (raw to processed) | Typically 30–90 minutes from overpass to geolocated detection, depending on ground-station contact and processing pipeline |
| Archive depth | HawkEye 360 commercial archive from 2019; Spire STRATOS from approximately 2021; Inmarsat interference advisories available on request from ITU filings |
| Cloud sensitivity | None. RF collection is unaffected by weather or atmospheric conditions |
| Coverage | Global, including ocean areas beyond coastal radar range; polar coverage included in Iridium and Spire orbital geometries |
Analytics Satellize can run
| Channel occupancy time-series | Power spectral density integration across defined L-band sub-bands per overpass, aggregated to hourly or daily occupancy percentage | Time-series CSV or GeoJSON feed, segmented by sub-band and geographic cell |
| Interference event detection alert | Threshold exceedance on carrier power or out-of-band emission detected during overpass; cross-referenced against ITU filing database for authorisation status | Timestamped alert with frequency, estimated location, and confidence tier; delivered via API or email within 90 minutes of overpass |
| Emitter geolocation report | TDOA/FDOA processing across multi-satellite intercepts; location history built from repeated detections over days to weeks | PDF report with location polygon, detection history chart, and vessel-AIS cross-reference where available |
| Congestion hotspot map | Spatial aggregation of occupancy detections over a user-defined period; kernel density estimation to identify persistent high-load areas | GIS layer (GeoTIFF or shapefile) suitable for overlay on maritime traffic charts |
| Regulatory evidence package | Compilation of raw intercept metadata, geolocation outputs, and ITU filing cross-reference into a structured evidentiary record | Structured PDF and supporting data archive formatted for submission to national maritime authority or ITU Radiocommunication Bureau |
| Baseline occupancy characterisation | Statistical analysis of archive data to establish normal occupancy distributions by sub-band, time of day, and geographic region; anomaly thresholds derived from baseline | Baseline report with anomaly detection parameters, updated quarterly or on request |
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.