04 · Earth-to-Orbit Comms

The downlink bottleneck and the LEO blackout.

A CubeSat can gather gigabytes of imagery and sensor data — but only matters if that data reaches the ground. In low Earth orbit, that turns out to be the hardest part. The satellite moves at roughly 7.5 km/s, sweeps over any given ground station in minutes, and then spends most of its orbit out of contact with anyone who can hear it.

5–15 min
Line-of-sight to a target ground station per orbit
~90 min
Time for one full LEO orbit
~3,000 km
Coverage radius of a single ground station
100–150 min
Worst-case store-and-forward delay to deliver a message
The geometry problem

Why contact is so brief

A satellite in low Earth orbit can only exchange data with a ground station when the two have a clear line of sight — when the satellite is above the station's local horizon. Because the satellite is moving so fast and orbits so low, that window lasts only 5 to 15 minutes per pass, and a given station might see the satellite just a handful of times a day. The rest of the orbit, the satellite is over oceans, deserts, ice or jungle where no one is listening.

Historic LEO data systems such as Orbcomm and Argos operated with roughly 20 ground stations worldwide, each covering about a 3,000 km radius. That still leaves enormous "white" areas — oceans and large parts of continents like Africa and Australia — where the network simply has no presence.

The classic workaround

Store-and-forward, and what it costs

When a satellite can't see a ground station, the standard solution is store-and-forward: the satellite records the data on board and carries it physically around the orbit until it flies over a station, then dumps it. It works, and it's robust — but it imposes latency measured in tens of minutes to well over an hour. For a remote sensor sending an occasional reading, that's tolerable. For anything that needs to be timely — disaster response, tracking, an emergency message from an off-grid traveler — a delay of up to 100–150 minutes is the difference between useful and useless.

This latency is the "LEO blackout" problem. The satellite isn't broken and the data isn't lost — but for most of every orbit there is no live path between the spacecraft and the ground. That gap is the bottleneck the OrbitRelay concept sets out to close.
What's changing

Inter-satellite links rewrite the rules

The biggest shift in the last few years is the rise of inter-satellite laser links in large constellations. Starlink, for example, connects its satellites to each other with a mesh of optical links that move data at the speed of light through vacuum. A satellite that connects to a user no longer needs its own line of sight to a ground station at that instant — it can pass the data satellite-to-satellite across the mesh until it reaches one that does.

That same architecture is what powers Starlink's Direct-to-Cell service, which extends ordinary LTE phone connectivity into regions with no cell towers. By 2025–2026 it had become the largest 4G coverage provider on Earth and was rolling out across markets in Africa and beyond. The lesson for small-satellite operators is clear: a dense relay layer in orbit, and a dense layer of user terminals on the ground, together can dissolve the coverage gaps that store-and-forward could only work around.

The OrbitRelay thesis: a constellation operator like Starlink already owns every leg of this tunnel — terminal, inter-satellite mesh and gateway — and already has terminals in the remote regions where LEO satellites go dark. Partner with that operator to relay a CubeSat's data through terminals on site, and the blackout starts to close itself. See the full proposal and infographic →

Continue

Sources: NTIA — Overview of LEO Satellite Systems · Practical horizon plane & LEO contact duration · Starlink Direct-to-Cell coverage (2025) · Direct Satellite-to-Device RAN — measurements (2025)