Radio astronomy detects some of the faintest signals in the universe. Faint echoes from cosmic hydrogen, the ticking of distant pulsars, and the signatures of magnetic fields of galaxies all reach us as radio waves.
Keeping the sky quiet enough to hear them is now harder than at any time in the history of the discipline. The reason is not a single new transmitter. It is the rise of large constellations of satellites in low Earth orbit – and the diversity of signals each satellite produces.
What we are picking up from orbit
Over the past three years, careful measurements with telescopes have catalogued thousands of detections of satellite signals, some in radio astronomy protected bands. Examples include findings from the Low Frequency Array (LOFAR) in Europe (documented by Di Vruno et al., 2023, and Bassa et al., 2024), the EDA2 prototype station for the future SKA-Low telescope in Western Australia (Grigg et al. 2023), the South Pole Telescope at millimetre wavelengths (Foster et al., 2025), and the Mopra radio telescope and Australia Telescope Compact Array in New South Wales (SNIFFLES-I survey).
The signals detected are not all alike. They originate in different parts of the satellite, are subject to different rules, and call for different responses.
Three kinds of signal
The ITU Radiocommunication Sector (ITU-R – one of three sectors of the International Telecommunication Union) employs the established vocabulary of the Radio Regulations to describe satellite signals. These descriptions are an important consideration in the context of the upcoming World Radiocommunication Conference (WRC-27).
Intended emissions are what a satellite is transmitting to fulfil its intended purposes. These include services such as Internet from satellites, satellite phone service, broadcasting services, or direct-to-cell services, noting the latter currently operates under national authorizations only, in derogation of the Radio Regulations. Such transmissions are powerful by design and closely regulated, with allocated frequencies, power limits, and, in some cases, coordination obligations toward radio astronomy stations.
The challenge for radio astronomy is the sheer power of the signals. Pointing a sensitive telescope anywhere near such a satellite can saturate the receiver outright.
Unwanted emissions are the residue of those intended emissions. No transmitter is perfectly clean. Out-of-band emission is the slope of the frequency content of a signal which extends just outside its assigned channel. Spurious emission appears further away in frequency, often as harmonics.
The SNIFFLES survey detected the second, third, and fourth order harmonics of a direct-to-cell carrier at approximately 5.25, 7.87, and 10.5 gigahertz (GHz). The latter one falls adjacent to a primary radio astronomy band used for continuum observations. Unwanted emission is not unregulated.
The relevant articles of the Radio Regulations and the spurious emission limits in Appendix 3 already apply. The policy task is enforcement and, where the limits are too generous for a passive science user, refinement.
Unintended radiation is different again. It may not come from a transmitter at all, but from the rest of the satellite. Switch-mode power supplies, clocks and oscillators, digital backplanes, motor controllers, and the inverters that condition the output of solar panels can all radiate at radio frequencies. Unlike unwanted emissions, these signals are not explicitly addressed in most spectrum management frameworks.
LOFAR observations identified broad and narrow band signals across 110 to 188 megahertz (MHz) from second-generation Starlink satellites at levels below what would be expected from any deliberate transmission, but still strong enough to potentially interfere with radio astronomy service operations in primary bands. The South Pole Telescope and the SKA-Low EDA2 prototype have measured similar emissions at low and high frequencies.
These signals are an electrical engineering by-product of a complex spacecraft rather than a feature of any radio service, but they are nonetheless emitted by a space station and can contribute to harmful interference to the radio astronomy service.
Why telling them apart matters
For unwanted emissions, the regulatory framework exists. However, it must be applied as rigorously as for terrestrial transmitters.
There is precedent for administrations addressing electromagnetic compatibility (EMC) issues that cause interference to radiocommunication services, such as the 1.4 GHz interference from TV receivers affecting the SMOS earth observation satellite, intended to measure soil moisture and ocean salinity (see report). However, these cases have been terrestrial in nature. For unintended radiation from satellites, there is at present no clear binding requirement on a satellite operator to prevent these from producing interference.
The internal electromagnetic compatibility standards used in the space industry, such as MIL-STD-461, AIAA S-121A, ECSS-E-ST-20-07C and the NASA SSP series, are written to ensure a satellite does not interfere with itself, with its co-passengers on the launch vehicle, or with its operator on the ground. They were not written to protect a radio telescope. Even AIAA S-121A’s space-science variant, which includes strict limits in passive-sensor bands, does so to protect the platform’s own instruments, not an observatory on the ground.
Yet the Radio Regulations establish that the obligation not to cause harmful interference to the radio astronomy service applies to a radiocommunication station as a whole, not only to the transmitter at its core. The legal foundation is in place; what is missing is implementation.
Toward WRC-27
The path forward is not adversarial. Some telecom operators have been an early and constructive partner, voluntarily characterizing satellites in anechoic chambers before launch and discussing on-orbit findings with radio astronomy operators. CSIRO (Australia’s national science agency), the SKA Observatory, the National Radio Astronomy Observatory, and other organizations are continuing similar conversations with operators, supported by international coordination through the IAU Centre for the Protection of the Dark and Quiet Sky (IAU CPS).
Two priorities for the WRC-27 cycle stand out:
- The existing rules on unwanted emission from satellites need to be applied with the same seriousness as for terrestrial transmitters, including in ongoing studies under WRC-27 agenda items 1.12 and 1.13 on possible new allocations to the mobile-satellite service (MSS).
- Work has begun, in cooperation between ITU-R, the International Special Committee on Radio Interference (CISPR), and the International Organization for Standardization (ISO), on practical electromagnetic compatibility limits for unintended radiation from spacecraft. Candidate per-satellite limits already exist: derived from measured satellite emissions and expressed in the field-strength conventions these bodies use, they scale with constellation size and are directly comparable to the radiated-emissions masks the industry already applies to itself (Indermuehle et al. 2026, PASA, submitted).
The technology to measure these emissions exists, and so does the legal foundation in the Radio Regulations. What is needed now is the will to bring them together, before the next generation of telescopes finds the radio sky a noisier place than the one their designers planned for.
