Spectrum & PolicyLeo SatellitesSpectrum SharingFccItu
LEO Operators Face Conflicting US and ITU Spectrum Rules
A UCLA paper finds over 10,000 LEO satellites face conflicting US and ITU interference rules, with real coexistence criteria unlikely before WRC-31.
Why it matters
- More than 10,000 commercial communications satellites operate in LEO, with 50,000+ more slated to launch over the coming decade.
- FCC rules cap later-round systems at 3% throughput degradation and 0.4% extra outage time — roughly 35 hours per year — for incumbents.
- ITU rules impose no quantified interference criteria for NGSO coexistence, relying on first-come filing priority outside the US.
- Satellite selection can preserve over 95% of link capacity while meeting regulatory interference thresholds, per cited research.
- No WRC-27 agenda item addresses NGSO sharing criteria; any push would likely come via WRC-31 agenda proposals.
The story
More than 10,000 commercial communications satellites now operate in low Earth orbit, with another 50,000-plus slated to launch over the coming decade — and no regulator has a consistent framework to keep their signals apart. That is the core finding of a paper published this month by Ian Roberts of the University of California, Los Angeles, which argues that LEO operators face a patchwork of interference rules that pulls them in opposite directions on either side of US borders.
The problem is structural. Unlike terrestrial mobile operators, satellite operators do not receive exclusive frequencies. They share spectrum, and their signals can clash. The paper gives a concrete example: Amazon Leo (formerly Kuiper) "operates user downlinks over the same frequencies that Starlink and OneWeb use for high-capacity gateway downlinks".
How do the two rulebooks differ?
The US and the ITU take fundamentally different approaches.
In the US, the FCC sets quantified interference limits:
- Systems authorised in the same FCC processing round share spectrum on equal terms, and interference at a receiver must stay below "roughly one-seventeenth of its noise floor".
- Systems from later rounds must protect earlier ones. Over time, a newcomer cannot reduce an older system's throughput by more than 3% or increase its outage time by more than 0.4% — which the paper calculates equals "roughly 35 hours per year".
The ITU, by contrast, has no equivalent yardstick. "The ITU does not formally quantify excessive interference for coexistence between NGSO constellations," the paper says. Priority goes to whichever filing reached the ITU first, and later systems must negotiate with earlier ones. The practical consequence: constellations that share spectrum equally over the US "must revert to strict first-come priority over international waters and foreign territories".
The paper also identifies a gap in the FCC's own regime. Each newcomer is assessed individually, so "two later-round constellations could each degrade an incumbent link by 3%", producing a combined loss above 6% while both remain compliant. The rules cap each interferer, not the cumulative total.
Can engineering fix coexistence?
On the technical side, Roberts argues that beamforming alone cannot solve interference. A rival satellite sometimes appears in almost the same direction as a user's own satellite, so the terminal cannot block one without blocking the other. The better tool, he writes, is satellite selection — handing users over to a satellite at a wider angle from the competing spacecraft.
The incentives align. "Steering clear of a competing satellite is in fact mutually beneficial, since it also shields the operator's own ground users from incoming interference," the paper says. Citing earlier research Roberts co-authored, the paper reports that this approach on its own "can satisfy regulatory interference thresholds across large-scale networks while preserving over 95% of link capacity".
Operator-to-operator cooperation offers a second route. The paper floats a centralised registry, similar to the databases used to share orbital data for collision avoidance, or private data links between pairs of operators. Roberts is candid about the commercial friction: "sharing operational schedules naturally introduces strategic and game-theoretic tensions". Operators in the same FCC round face an even harder problem. "Neither operator holds spectrum priority, creating a deadlock over who yields unless mandatory band-splitting takes effect."
Why is the pressure increasing?
Roberts expects interference problems to worsen as the number of operators grows. "The default regulatory remedy of mandatory equal band-splitting, for instance, degrades rapidly as the number of operators grows," he writes.
Direct-to-cell systems will add a further layer of congestion. Their links to phones use frequencies below 3 GHz, but they backhaul traffic to the ground in the same Ka- and Q/V-band allocations used by broadband constellations. The paper cites a July 2026 Amazon Leo filing for a 5,105-satellite direct-to-cell constellation as evidence of the coming overlap.
What happens at WRC-27?
The paper's main policy call targets the ITU. Following the FCC's shift to performance-based limits, it argues that "establishing comparable criteria within the ITU is needed to accommodate coexistence mechanisms and support spectrum sharing as mega-constellations expand".
That will not happen at the next World Radiocommunication Conference. None of the WRC-27 agenda items sets criteria for sharing between NGSO constellations, and the satellite procedures item — agenda item 7 — covers unrelated topics. ITU working parties have now finalised their draft texts for the Conference Preparatory Meeting report, which goes to the second CPM session in April 2027 ahead of WRC-27 in Shanghai.
The realistic route, the paper concludes, is slower: any push for NGSO coexistence criteria would most likely come through proposals for the WRC-31 agenda.
Also reported
Original: arxiv.org
More from Daniel Okafor
Show full bio
Senior reporter covering regulation, policy and their impact on Carrier business, 5G and 6G, fiber and broadband at Telecom Gazette.
540 articles