LORAN and Chayka navigation transmitter monitoring from orbit
LEO RF payloads can detect 90–110 kHz LF skywave emissions from LORAN-C and Chayka stations, revealing operational status, pulse-timing drift, and unexpected activations that ground-based monitoring networks miss.
Sensors
- HawkEye 360 Cluster Satellites: Tri-satellite formation flying in LEO at roughly 575 km altitude. Time-difference-of-arrival and frequency-difference-of-arrival processing across the cluster enables geolocation of emitters. Publicly stated geolocation accuracy is in the range of hundreds of metres to a few kilometres depending on geometry and signal duration. The LF band capability is constrained by antenna aperture on a small satellite, but strong LORAN-C pulse groups (peak radiated power at major stations reaches hundreds of kilowatts) are detectable as skywave returns.
- Spire LEMUR-2 RF Payload: LEMUR-2 satellites carry a software-defined radio payload covering a broad frequency range. The constellation exceeds 100 satellites in LEO, giving revisit intervals of roughly one to three hours at mid-latitudes. The SDR architecture allows post-processing tuning to the 90–110 kHz LF band, though sensitivity at these frequencies from a small platform is a genuine constraint. Useful primarily for detecting high-power stations and confirming presence or absence of emission rather than fine timing analysis.
- Published LORAN-C / Chayka Monitoring Literature (US Coast Guard, IALA): Not a sensor, but the methodological baseline. US Coast Guard and IALA technical reports document the LORAN-C pulse group format, group repetition intervals (GRIs ranging from roughly 49,900 to 99,900 microseconds), and the signal-in-space requirements against which any space-based measurement must be validated. These publications establish what 'nominal' looks like, which is what makes anomaly detection possible.
- Ionospheric Monitoring Receivers (e.g., aboard COSMIC-2 / FORMOSAT-7): Although designed for GNSS radio occultation, LF propagation modelling depends on ionospheric electron density profiles. COSMIC-2 provides near-real-time ionospheric data that can be used to correct for skywave propagation delays, improving the reliability of timing-integrity assessments made from orbit.
Why anyone still cares about a 1950s navigation system
LORAN-C and its Russian counterpart Chayka are pulsed, ground-wave and skywave navigation systems operating in the 90–110 kHz low-frequency band. Both were declared largely obsolete after GPS became ubiquitous. Several countries decommissioned their transmitter chains between 2010 and 2015. Yet interest has revived sharply, for one reason: GPS can be jammed or spoofed with cheap equipment, and LF navigation cannot. South Korea reactivated an eLORAN chain after repeated GPS jamming incidents near the North Korean border. The United Kingdom funded an eLORAN trial. The US Congress has repeatedly directed the Department of Transportation to maintain a backup capability.
The practical consequence is that a small number of LORAN-C and Chayka stations remain operationally significant, and several more exist in ambiguous states: nominally decommissioned but physically intact, or operated intermittently by military users. Knowing which stations are actually transmitting, at what power, and with what timing integrity, is a legitimate intelligence and navigation-safety question. Ground-based monitoring networks answer it well inside their coverage footprint. Outside that footprint, space-based RF monitoring is one of the few independent verification methods available.
What a LEO receiver actually picks up at 100 kHz
LF signals do not propagate to orbit the way VHF or UHF emissions do. The mechanism is ionospheric skywave reflection. A LORAN-C pulse group radiated from a high-power station (typically 250–1,000 kW effective radiated power at major stations) propagates upward, reflects off the D and E layers of the ionosphere at roughly 70–90 km altitude, and a fraction of that energy continues outward. A LEO satellite at 500–600 km altitude sits above the ionosphere and intercepts the upward-propagating component before it reflects back down. The signal is attenuated, but the pulse structure, carrier frequency, and group repetition interval are preserved well enough for identification.
This is not a theoretical claim. The ionospheric physics of LF propagation at these altitudes is well documented in the academic literature on VLF/LF communications with submarines, where signals routinely couple into and out of the ionospheric waveguide. The challenge for a small satellite is antenna efficiency at 100 kHz: a quarter-wave antenna at that frequency would be 750 metres long. In practice, electrically short antennas with significant impedance mismatch are used, and the system depends on the very high transmit power of the ground stations to compensate.
What can be determined, and what cannot
A space-based receiver can determine, with reasonable confidence, whether a station is transmitting, which GRI it is using (and therefore which chain it belongs to), and whether the carrier frequency is close to nominal. Pulse group timing can be extracted, though the accuracy achievable from orbit is degraded by ionospheric propagation variability, which introduces timing jitter that ground-based monitors correct for using well-characterised propagation models. From orbit, that correction is harder because the propagation path geometry changes continuously as the satellite moves.
Spatial resolution is the most significant honest limit. HawkEye 360's TDOA geolocation of a continuous or long-duration emitter achieves kilometre-class accuracy under good geometry. LORAN-C pulse groups are short (roughly 250 microseconds per pulse), which reduces the effective integration time available for TDOA. Locating a known station to within a few kilometres is feasible; locating an unknown emitter to the accuracy needed to distinguish two nearby stations is much harder. Co-channel separation is the other hard problem. Multiple Chayka stations share the same carrier frequency and are distinguished solely by their GRI. If two stations with different GRIs are both in the satellite's field of view, their pulse groups interleave and can in principle be separated by GRI analysis, but this requires clean signal capture and careful processing. Stations with the same GRI cannot be separated by timing alone from a single satellite pass.
Operational status monitoring as the primary product
The most defensible application is binary: is this transmitter on or off? For a decommissioned station that should be silent, any detected emission at the correct frequency and GRI is an anomaly worth reporting. For an active station, absence of detection across multiple passes is evidence of outage, though the satellite's sensitivity limits mean a single non-detection is not conclusive.
Repeated passes build a statistical picture. With Spire's constellation at current size, a given point on Earth receives multiple revisits per day. A station that is consistently detected across ten passes and then absent for three consecutive passes has almost certainly gone off-air. That kind of time-series is actionable for maritime safety authorities, eLORAN service planners, or defence customers assessing the operational state of a foreign navigation infrastructure. It is not a substitute for a dedicated ground-based monitoring receiver near the station, but it is independent, non-cooperative, and globally available.
The Chayka dimension: asymmetric information
Russian Chayka stations use the same 100 kHz carrier and a compatible pulse format to LORAN-C, with GRIs assigned to distinguish chains. During the Cold War, the two systems were designed to be interoperable under certain conditions. Several Chayka stations are located in areas where ground-based Western monitoring is limited: the Russian Arctic, Siberia, and the Russian Far East. Space-based monitoring is one of the few methods available to an external observer wanting to assess whether these stations are active, whether their GRIs have changed, or whether new emissions have appeared at LF frequencies in those regions.
This is where the honest limits matter most. Detecting emission from a Chayka station in the Russian Arctic from a single LEO pass is plausible given the signal power involved. Characterising its timing integrity to the level needed for navigation use is not realistic from orbit alone. The space-based product here is strategic awareness, not precision navigation validation. Satellize treats this distinction as non-negotiable in how it frames deliverables to clients.
Building a monitoring programme around sparse data
No single commercial constellation was designed for LF monitoring. The practical approach is to combine HawkEye 360 tasking for geolocation events with Spire SDR data for time-series presence detection, then apply propagation modelling to interpret what the receivers see. Ionospheric state data from COSMIC-2 or similar sources corrects for the day/night asymmetry in LF skywave propagation, which is substantial: nighttime skywave propagation is significantly stronger than daytime because the D-layer absorption that attenuates LF signals during the day largely disappears after sunset.
The output is a monitoring report structured around individual stations or chains, updated on a cadence matched to the client's need. For a maritime authority assessing eLORAN service continuity, weekly summaries may suffice. For a defence customer tracking a specific Chayka chain, near-real-time alerting on detected state changes is more appropriate. Archive depth for commercial RF data from HawkEye 360 and Spire extends back several years for some coverage areas, allowing historical baselining before a monitoring contract begins. Satellize applies the same analytics pipeline it uses for its Tonga crop-estimation programme to this problem: systematic, documented, and honest about what the data cannot prove. If you want to scope a monitoring package for a specific station list, the right first step is a feasibility review against the published coverage and sensitivity of the available constellations.
Typical figures
| Frequency band monitored | 90–110 kHz (LF); LORAN-C and Chayka carrier at 100 kHz |
| Emitter geolocation accuracy | Kilometres-class for known-format emitters via TDOA (HawkEye 360); insufficient for sub-kilometre discrimination of co-located stations |
| Revisit interval | 1–3 hours at mid-latitudes (Spire LEMUR-2 constellation); lower frequency for HawkEye 360 tasked passes |
| Minimum detectable transmit power | Practically limited to high-power LF stations (hundreds of kW ERP); low-power test transmitters not reliably detectable from LEO |
| Timing integrity assessment accuracy | Degraded relative to ground-based monitoring due to ionospheric propagation jitter; useful for anomaly detection, not precision navigation validation |
| Co-channel station separation | Possible by GRI analysis when GRIs differ; not feasible for same-GRI stations from a single pass |
| Archive depth | Several years for HawkEye 360 and Spire in covered areas; station-specific availability must be confirmed |
| Ionospheric correction data | COSMIC-2 electron density profiles used to model day/night propagation asymmetry in LF skywave |
| Delivery format | Station status reports (PDF/JSON), time-series presence logs (CSV), alert feeds for state-change events |
Analytics Satellize can run
| Transmitter on/off status per station | Presence/absence detection across repeated LEO passes; threshold set against known station ERP and modelled skywave path loss | Weekly or near-real-time station status report per chain, with confidence rating per observation |
| GRI identification and chain attribution | Pulse group repetition interval extraction from captured LF waveform; matched against published LORAN-C and Chayka GRI tables | Chain attribution table appended to each detection event log |
| Anomalous activation alert | Comparison of detected emissions against baseline of expected operational stations; flagging of emissions from nominally decommissioned sites | Alert notification with pass time, frequency, GRI, and estimated bearing or geolocation polygon |
| Timing anomaly indicator | Pulse group timing extracted from SDR capture; compared against nominal GRI with ionospheric propagation correction applied; deviations flagged | Time-series plot of measured vs nominal pulse timing per station; anomaly flags in accompanying JSON feed |
| Historical baseline assessment | Retrospective analysis of archived RF data over selected stations to establish operational patterns before monitoring contract begins | Baseline report covering detected activity periods, outages, and GRI consistency over archive period |
| Chayka chain operational assessment | Multi-pass aggregation of detection events for Russian Arctic and Far East stations; cross-referenced against ionospheric state to account for propagation variability | Quarterly strategic assessment report on Chayka chain operational status with confidence intervals |
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.