Domain 05 · Space
Nothing up there is a sanctuary
Custody of every object that matters. A vehicle that repositions in days. A defence that produces no debris. And a position when GPS is gone.
The thesis
Space stopped being a sanctuary the first time one satellite moved next to another on purpose
Everything since has been about the fact that we cannot tell why.
For sixty years the orbital environment was a place you put things and left them. Orbits were predictable, populations were small, and the only realistic threat was a ground-launched interceptor that everybody agreed nobody would use. Catalogue maintenance was a bookkeeping job. It is not a bookkeeping job any more.
Three things changed at once. The active population went from roughly a thousand satellites to more than ten thousand in fifteen years, so the sky is now crowded enough to hide in. Electric propulsion made routine, low-signature manoeuvre affordable, so objects no longer stay where the propagator says they will be. And several states demonstrated — publicly, repeatedly, and without violating any treaty — that a spacecraft can be flown up to somebody else's spacecraft and parked there.
The consequence is a specific and unpleasant intelligence problem. For the first ninety-five per cent of the trajectory, an inspection, a servicing mission, a rescue and an attack are the same manoeuvre. Nothing observable separates them. The United States has world-class detection and effectively no custody: we find things, and then we lose them, and then we find them again and cannot prove they are the same thing doing the same job. Without an unbroken record of an object behaving normally, the word "abnormal" has no meaning.
Underneath that sits a harder constraint. The kinetic answer is closed. The 2007 and 2021 destructive intercepts each put over a thousand catalogued fragments into orbits that we ourselves use, and the United States renounced destructive direct-ascent testing in 2022. So the response has to be non-kinetic, it has to be proportionate, and — because everything in this domain is invisible and deniable — it has to be provable. A defensive action nobody can verify is not a deterrent. It is a second ambiguity stacked on the first.
And the whole architecture rests on a positioning signal broadcast from 20,200 km at roughly one ten-thousandth of a trillionth of a watt. GPS jamming and spoofing across the Baltic, the Black Sea and the Eastern Mediterranean is now a daily civil-aviation fact. Any plan that assumes GPS is a plan with a single point of failure that the adversary already knows how to press.
Four requirements fall out of that, and they are the four systems on this page.
The public record
None of this is our claim. All of it is on the record.
| 2007 | China destroys Fengyun-1C at 865 km. More than 3,000 catalogued fragments, most of them still on orbit today. |
|---|---|
| 2021 | Russia destroys Cosmos-1408 with a Nudol interceptor. Roughly 1,500 catalogued fragments; the ISS crew shelters in their return vehicles. |
| 2022 | The United States unilaterally renounces destructive direct-ascent anti-satellite testing. The kinetic response is now closed to us by policy, and nothing was fielded to replace it. |
| 2014 — | A Russian Luch/Olymp spacecraft drifts the geostationary belt, parking within tens of kilometres of Western communications satellites. Legal. Deniable. Repeated. |
|---|---|
| 2020 | Kosmos-2542/2543 shadows a US satellite in low Earth orbit, releases a sub-satellite, and later ejects a high-speed projectile. US Space Command publicly characterises it as a non-destructive anti-satellite test. |
| 2022 | China's Shijian-21 docks with a defunct BeiDou-2 satellite and tows it into a graveyard orbit. That capability has no benign version and no hostile version. It is one capability. |
| 2022 — | Persistent GNSS jamming and spoofing across the Baltic, Black Sea and Eastern Mediterranean, affecting civil aviation daily. |
Every event above is drawn from public government statements and open catalogue data. Fantom Defense publishes no classified or export-controlled information on this site.
Four requirements, four systems
The space arsenal
Space domain key figures
One architecture
These are not four products
Read this part
What we do not claim
- Nothing in this domain has flown. LANTERN has a manifested pathfinder pair. CHARON is in propulsion qualification. SHROUD is a ground demonstration. BEACON is a signal definition. Every performance figure on these pages is a design requirement, not a measurement.
- LANTERN's solar exclusion cone is a published, predictable volume. A competent adversary can compute exactly when their object sits inside it.
- A dark, cold, radio-silent object in Earth's shadow is held by neither the optical payload nor the RF payload. That is the cleanest hole in the architecture and we have not closed it.
- CHARON cannot reach GEO from a 28.5° low Earth parking orbit. It needs a geostationary transfer drop-off. Large plane changes in LEO are unaffordable at any specific impulse anyone can fly.
- SHROUD's illuminator will be called a weapon regardless of the hardware interlock that holds it three orders of magnitude below sunlight.
- We are proposing a norm of accountable proximity operations, not citing one. Outer Space Treaty Article IX has no enforcement mechanism, and a cryptographic record makes lying expensive rather than impossible.
- BEACON is a receiver programme as much as a satellite programme. Every GPS receiver in the US inventory is worthless for it, and the installed base dwarfs the space segment in cost and schedule.
- We are not the first to propose LEO alternative PNT. Iridium's time-and-location service has been fielded for over a decade, and at least one commercial constellation is building now.
Full parameter tables, worked orbital mechanics and the complete open-risk list for every system are on the product pages.
Space programs
Bring us the orbit you cannot cover
Give us the regime, the revisit you need, the manoeuvre threshold you have to detect and the time you have to respond in. We will tell you what it costs, what it weighs, and which part of it we cannot do yet.