DAB and DVB-T digital broadcast signal monitoring from orbit
Spaceborne RF receivers can detect, geolocate and fingerprint DAB and DVB-T transmitters by exploiting their OFDM pilot-tone structure, enabling regulators to verify compliance, resolve cross-border interference disputes and identify unlicensed transmitters without ground-based inspection.
Sensors
- HawkEye 360 cluster satellites: Three-satellite formation flying in LEO at roughly 575 km altitude. TDOA/FDOA geolocation of emitters across a wide frequency range covering VHF and UHF, where DAB (174–240 MHz) and DVB-T (470–790 MHz in most ITU Region 1 allocations) both sit. Published geolocation accuracy is typically 1–5 km CEP depending on geometry and signal duration; revisit at any given point is multiple times per day across the full constellation.
- Spire STRATOS RF payload: Spire's STRATOS software-defined radio payload hosted on Lemur-2 satellites collects wideband IQ data in LEO. The SDR architecture allows post-collection demodulation of OFDM signals including DAB and DVB-T frames, enabling pilot-tone extraction and ensemble/multiplex identification from recorded captures. Revisit depends on constellation density and tasking priority.
- ITU/NOAA interference monitoring filings: Not a spaceborne sensor, but the ITU Radio Regulations and associated Radiocommunication Bureau filings define the licensed frequency assignments, ERP limits and antenna patterns against which spaceborne measurements are compared. NOAA coordinates US positions within ITU processes. These filings are the regulatory ground truth for compliance verification.
- Generic wideband SDR cubesats (commercial and experimental): Several commercial operators fly wideband SDR payloads capable of capturing UHF spectrum. Geolocation from a single satellite requires either Doppler-only methods (accuracy typically 5–20 km, depending on pass geometry and signal duration) or hand-off to a multi-satellite TDOA architecture. Single-pass Doppler is sufficient for detection and rough attribution; precise geolocation requires cluster geometry or multi-pass fusion.
What OFDM pilot tones give away from 575 km
DAB and DVB-T are both built on orthogonal frequency-division multiplexing. Both standards embed continuous and scattered pilot carriers at defined subcarrier positions and known relative power levels. From a spaceborne receiver, these pilots are not a vulnerability in any security sense, but they are a gift to the analyst: they allow a receiver with no prior knowledge of the transmitted content to identify the signal type, measure the carrier frequency offset caused by Doppler, and extract the ensemble or multiplex identifier encoded in the transmission parameter signalling (TPS) pilots of DVB-T or the Fast Information Channel of DAB.
A LEO satellite passing at roughly 7.5 km/s induces a Doppler shift of several kilohertz on a UHF carrier. That shift is predictable and can be removed in post-processing once the satellite's precise ephemeris is known. What remains is the clean OFDM spectrum of the transmitter below, from which pilot positions, guard interval length and mode can be read directly. DVB-T mode 8K, for instance, places 6,817 active subcarriers across an 8 MHz channel with a 1/32 to 1/4 guard interval; those structural features survive the Doppler correction and are unambiguous.
Geolocation geometry: what TDOA and FDOA can and cannot resolve
Time-difference-of-arrival between satellites in a formation like HawkEye 360's three-satellite cluster places an emitter on a hyperboloid surface. Frequency-difference-of-arrival adds a second constraint surface. The intersection of two such surfaces, combined with the assumption that the emitter is on the Earth's surface, typically yields a geolocation ellipse. Published figures for HawkEye 360 suggest 1–5 km CEP for cooperative, continuous emitters in good geometry. A DAB transmitter running at 10 kW ERP is well above the detection floor; a low-power gap-filler or single-frequency network relay at 100 W or less is harder and may require longer dwell or multiple passes.
Single-satellite Doppler-only geolocation is considerably less precise, typically 5–20 km depending on pass elevation angle and signal duration. It is adequate for detection and country-level attribution but not for locating a transmitter to a specific tower. Cross-border interference disputes, where the question is which country's transmitter is the source, generally need the better accuracy of a multi-satellite TDOA solution. Honest caveat: if two transmitters in a single-frequency network are transmitting on the same frequency within the OFDM guard interval window, a spaceborne receiver may see them as one emitter. Separating SFN nodes requires either very high time resolution or multiple passes from different geometries.
Regulatory compliance: measuring what the licence actually permits
An ITU filing for a DAB ensemble or DVB-T multiplex specifies the carrier frequency, channel bandwidth, effective radiated power, antenna height and horizontal radiation pattern. A spaceborne measurement cannot directly verify the antenna pattern, but it can measure the received power flux density at the satellite, and with a known link budget (transmitter-to-satellite path loss, satellite antenna gain, receiver noise figure) it can back-calculate the ERP in the direction of the satellite. Repeated passes at different azimuth angles build a partial picture of the radiation pattern.
Where a transmitter is operating significantly above its licensed ERP, the excess shows up clearly. Where it is transmitting on a frequency adjacent to or overlapping a licensed channel, the OFDM spectrum captured at the satellite shows the spectral mask violation directly. This kind of measurement is not a replacement for a ground-based field-strength survey under ITU-R Recommendation BS.704 or similar, but it provides an independent, tamper-resistant observation that a regulator can use to initiate an investigation or support a dispute filing with the Radiocommunication Bureau.
Clandestine transmitters: the multiplex impersonation problem
A clandestine transmitter mimicking a licensed DVB-T multiplex is a specific threat that has appeared in several documented interference cases in Europe and the Middle East. The attacker transmits on the licensed frequency with content that may include disinformation or propaganda, relying on the receiver's automatic network scanning to treat the signal as legitimate. From the ground, the interference may be intermittent and hard to locate. From orbit, the transmitter's presence is visible whenever a satellite is in view, and its geolocation can be attempted on each pass.
The key discriminant is the TPS cell identifier embedded in the DVB-T signal. A legitimate transmitter in a licensed network carries the cell ID assigned by the network operator. A clandestine transmitter either omits it, copies an existing ID (which creates a detectable ambiguity when both are in view), or uses an unregistered value. Spaceborne IQ capture with sufficient bandwidth to demodulate the full DVB-T frame can extract this field. DAB has an analogous mechanism in the ensemble label and service component identifiers in the Fast Information Channel. Neither method is foolproof if the attacker has detailed knowledge of the legitimate signal structure, but most clandestine operators do not go to that level of engineering effort.
Practical limits a buyer should understand before commissioning a survey
Spaceborne RF monitoring of broadcast signals is not continuous surveillance. A LEO satellite passes a fixed point for a few minutes per orbit. A constellation like HawkEye 360's provides multiple passes per day, but a transmitter that operates only during specific hours, or that is switched off when a satellite is in view, may be missed. Persistent monitoring requires either a large constellation or a geostationary RF receiver, and no commercial GEO RF monitoring asset currently offers the sensitivity needed to detect a 10 kW UHF terrestrial transmitter reliably.
Urban environments create multipath that degrades TDOA accuracy. The 470–790 MHz DVB-T band in ITU Region 1 is congested, particularly in Western Europe, and separating overlapping signals from different transmitters in the same channel requires careful spectral analysis. Archive depth for commercial RF collections is typically months to a few years, not decades, so historical trend analysis is limited. Satellize's role in this use case is to commission collection from the appropriate RF constellation, apply the OFDM signal analysis pipeline, and deliver geolocated emitter reports to the client, in the same way it runs analytics pipelines on open optical constellations for clients such as the Kingdom of Tonga crop-estimation programme.
Typical figures
| Frequency coverage (DAB) | 174–240 MHz (VHF Band III); L-band DAB at 1452–1492 MHz in some markets |
| Frequency coverage (DVB-T/T2) | 470–790 MHz (UHF) in ITU Region 1; 470–698 MHz post-digital-dividend repack |
| Geolocation accuracy (TDOA/FDOA, multi-satellite) | 1–5 km CEP for continuous emitters above ~100 W ERP in good pass geometry (HawkEye 360 published figures) |
| Geolocation accuracy (single-satellite Doppler) | 5–20 km, depending on pass elevation angle and signal duration; adequate for detection, not for tower-level attribution |
| Minimum detectable ERP (indicative) | Signal detectability from LEO is geometry-dependent; 10 kW ERP transmitters are well above threshold; sub-100 W gap-fillers require favourable geometry or multiple-pass integration |
| Revisit (HawkEye 360 full constellation) | Multiple passes per day over most latitudes; exact revisit varies with orbital geometry and constellation size |
| Signal features extractable | Carrier frequency, channel bandwidth, OFDM mode and guard interval, DVB-T TPS cell ID, DAB ensemble/service identifiers, ERP estimate (via link budget) |
| Archive depth (commercial RF collections) | Typically months to a few years; not comparable to optical archive depth |
| Delivery latency | Hours to days from collection to geolocated emitter report, depending on downlink scheduling and processing pipeline |
Analytics Satellize can run
| Geolocated emitter catalogue | TDOA/FDOA multi-satellite geolocation applied to detected DAB/DVB-T carriers; emitter positions expressed as geolocation ellipses with CEP estimates | GIS layer (GeoJSON or shapefile) of detected transmitters with frequency, estimated ERP, geolocation uncertainty ellipse and pass timestamps |
| Licence compliance report | Measured ERP and carrier frequency compared against ITU filing database entries; spectral mask analysis from captured IQ data | PDF compliance report per transmitter, flagging frequency offset, ERP excess or spectral mask violations with supporting spectrogram evidence |
| Cross-border interference attribution | Multi-pass TDOA geolocation fused with ITU coordination zone boundaries and filed coordination agreements; signal identity confirmed via OFDM structural analysis | Interference attribution dossier suitable for submission to ITU Radiocommunication Bureau, including geolocation evidence, signal captures and link-budget calculations |
| Clandestine transmitter alert | DVB-T TPS cell ID and DAB FIC ensemble label extraction; comparison against licensed network database; anomaly flagging for unregistered or duplicated identifiers | Alert feed (JSON or email) triggered when an unregistered or conflicting multiplex identifier is detected, with geolocation estimate and signal capture excerpt |
| Temporal activity profile | Multi-pass detection log aggregated over days to weeks; transmitter on/off schedule inferred from pass-by-pass detection presence or absence | Time-series chart and CSV of transmitter activity windows, useful for identifying intermittent or schedule-based clandestine operation |
| SFN node separation analysis | High-time-resolution IQ capture analysed for guard-interval echoes and multi-source OFDM arrival timing; where geometry permits, individual SFN nodes distinguished by arrival delay | SFN topology map showing probable node locations and relative signal strengths, with explicit uncertainty bounds where node separation is below resolution threshold |
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.