Inter-Satellite Laser Links

The problem they solve

Without inter-satellite links, every user packet has to go: user → satellite → ground station → internet. The satellite must be in view of both the user and a ground station simultaneously. For LEO satellites passing overhead, this is only true some fraction of the time. Coverage gaps occur over oceans, polar regions, and any area far from ground stations.

With inter-satellite links, the path becomes: user → satellite → (other satellites) → satellite over a ground station → ground station → internet. The constellation routes the data in space; only one satellite needs to be over a ground station.

Why optical instead of radio

In vacuum, optical links have no downsides relative to radio. In atmosphere, optical links suffer absorption, scattering, and weather effects — which is why satellite-to-ground links use radio while satellite-to-satellite uses optical.

Pointing precision

A laser beam from a LEO satellite to another LEO satellite ~1000 km away spreads to perhaps a few meters at the target. The pointing accuracy required is sub-microradian — about the angle subtended by a coin viewed from across the United States.

How they achieve it:

1. Coarse pointing based on orbital ephemeris (both satellites know where they are and where the other should be).
2. Beacon acquisition — each terminal emits a wider beacon that the other can detect.
3. Fine pointing — closed-loop tracking using the received beacon refines the pointing.
4. Lock — once aligned, the data laser is enabled and the link carries traffic.

Acquisition can take seconds. Once locked, the link stays locked for minutes or until orbital geometry breaks it.

The constellation routing problem

With inter-satellite links, a constellation becomes a routed network in space. Each satellite is a router; the links to its neighbors are dynamic links that come and go as orbital geometry changes. Path selection becomes a real-time routing problem:

• Forward routing: which path of satellites carries this packet to the destination ground station with lowest latency?
• Backup paths: what if a link fails or a satellite's laser is dazzled by the sun?
• Handoffs: as satellites move, the optimal next hop changes.

Constellations use both pre-computed routes (deterministic given orbital mechanics) and dynamic adjustments for failures.

Sun and bright object avoidance

Laser receivers are very sensitive — single-photon detectors are common. The sun is overwhelming. If the laser path between two satellites passes near the sun (a "sun outage"), the receiver is saturated and the link drops momentarily.

Sun outages are predictable from orbital geometry; constellations plan around them by routing through alternative paths during the brief seasonal windows where solar geometry interferes.

Eye safety

Laser links in space don't directly threaten ground observers — the beams are aimed at other satellites, not at Earth. But there are concerns about reflections during alignment and edge cases involving accidental ground exposure. Operators use eye-safe wavelengths and power levels where possible, and ground-based observatories coordinate with constellation operators to avoid laser interference with astronomical observations.

Bandwidth in practice

Current commercial inter-satellite laser links carry 10-200+ Gbps per link. Multiple links per satellite enable mesh topologies; a single satellite can connect to 4-6 neighbors simultaneously. Aggregate constellation bandwidth scales accordingly.

Compared to terrestrial fiber (which can carry tens of Tbps per fiber), laser links are smaller but adequate for distributing per-user satellite-internet traffic across the constellation.

Comparison to bent-pipe

A "bent-pipe" satellite simply relays incoming radio signal to a ground station — no on-board routing. Every packet must hit a satellite over a ground station. Coverage is limited to areas with a satellite-and-station in view.

With inter-satellite links, the constellation becomes "regenerative" — it routes intelligently. Coverage expands to areas without nearby ground stations. The complexity is higher; the capability is fundamentally different.

Real-world deployment

• Starlink Gen 2+ — laser links between satellites; routing over the constellation reduces dependence on ground stations.
• Iridium NEXT — radio inter-satellite links (not optical), but the same architectural pattern.
• OneWeb, Kuiper, others — planning or deploying optical inter-satellite links.

The trend is toward fully-meshed optical constellations where every satellite has multiple laser links to neighbors.

Future directions

• Direct satellite-to-ground laser links for ultra-high-bandwidth ground stations (when weather permits).
• Direct laser links from satellite to user terminal (research, not commercial yet).
• Optical mesh including different orbit shells (LEO-to-MEO laser links).
• Higher per-link bandwidths as optical component costs continue to drop.

Frequently Asked Questions

What is an inter-satellite laser link?

A free-space optical link between two satellites in orbit, using lasers instead of radio frequencies. Each satellite has a precision-pointing telescope that aims a laser beam at a neighboring satellite, and a receiver that captures the incoming beam. Used to route traffic between satellites in a constellation without needing every satellite to be in view of a ground station.

Why use lasers instead of radio?

Three reasons. Bandwidth: optical frequencies (hundreds of THz) can carry orders of magnitude more data than radio. Power efficiency: a tightly-focused laser beam delivers signal to the receiver with minimal power loss, unlike radio that spreads broadly. No regulatory constraints: laser links don't use ITU-allocated radio spectrum so there's no frequency coordination. In space (above the atmosphere), the disadvantages of optical links — atmospheric absorption, weather — don't apply.

How do satellites point lasers at each other?

Very precisely. The pointing requirement is sub-microradian — comparable to hitting a coin at the distance of New York from Los Angeles. Each terminal has a fine-pointing mechanism (gimbals, mirrors) with closed-loop control that uses a beacon signal from the target satellite. Acquisition starts coarse (based on orbital ephemeris) and refines via beacon tracking until the data laser locks on.

What does this enable that radio constellations couldn't?

Routing entirely in space. Without inter-satellite links, every satellite must be in view of a ground station to relay user traffic to the broader internet. This requires many ground stations and limits coverage in remote areas (oceans, polar regions). With laser links, traffic hops satellite-to-satellite until reaching a satellite over a ground station — coverage anywhere on the planet without dense ground infrastructure.

What are the disadvantages?

Cost and complexity. Optical terminals are expensive (precision optics, single-photon detectors, fine pointing mechanisms). Acquisition and re-acquisition takes time. Failures cascade — if one satellite's laser fails, multiple hops may need rerouting. And laser-to-laser links between non-adjacent satellites require either intermediate hops or much more complex pointing across longer distances.