hckrnws

Here's a paper (from July 2025) on previous steps in this program, getting up the initial testing in flight. Maximum uplink laser power of 20W, though they got good performance all the way down to 2W. The sat has a laser pointing down that was used to help lock on, but it's not clear if it has any meaningful downlink capability, all discussions are about uplink capability. Lots a nerdy details here.

https://www.spiedigitallibrary.org/conference-proceedings-of...

In addition, here's a random paper on the testing performed on the space borne laser terminals - https://icsos2012.nict.go.jp/pdf/1569586689.pdf

This tells us that the laser terminals have a FOV of +/-2.5mrad in acquisition mode (so before lock on), and +/-0.5mrad in communication/tracking mode. This corresponds ~100km and ~20km radius FOV from GEO to surface.

Uplink alone can be significant for clandestine operations.

You can have a stealth drone which is effectively invisible to SIGINT transmitting real time intelligence to a satellite whilst either operating autonomously or receiving commands via a wide area encrypted broadcast (yes I know you can theoretically detect receivers through signal attenuation but at these distances it’s effectively impossible to do).

> yes I know you can theoretically detect receivers through signal attenuation but at these distances it’s effectively impossible to do

Well if that's part of your threat model then you should also consider the RF put out by the motors. Remember the part where we densely blanked the inhabited parts of the world with highly sensitive antennas over the past 3 decades?

Yeah, spark of engine is something what can be seen on SIGINT for dozens of kilometers.

So unless jet engine is used (which rarely is on reconnaissance drones) then the drone will be spark lighthouse in the sky.

They say "error-free connection", which implies 2-way communication, right?

"low-latency links", says the article. I wonder if they consider 500 ms ping to be low, or if they want to replace Geostationary with Low Earth Orbit.

> "low-latency links", says the article. I wonder if they consider 500 ms ping to be low, or if they want to replace Geostationary with Low Earth Orbit.

Directional laser beams are orders of magnitude to jam compared to radio wave. That alone makes it of big interest for military applications, even with 500 ms latency.

There is several known cases where jamming caused the loss of costly military drones.

https://en.wikipedia.org/wiki/Iran%E2%80%93U.S._RQ-170_incid...

Laser comms could prevent that entirely.

> Directional laser beams are orders of magnitude to jam compared to radio wave. That alone makes it of big interest for military applications, even with 500 ms latency.

I am reminded of RFC 1217 - Memo from the Consortium for Slow Commotion Research (CSCR) https://www.rfc-editor.org/rfc/rfc1217

    2. Jam-Resistant Land Mobile Communications

       This system uses a highly redundant optical communication technique
       to achieve ultra-low, ultra-robust transmission.  The basic unit is
       the M1A1 tank.  Each tank is labelled with the number 0 or 1 painted
       four feet high on the tank turret in yellow, day-glo luminescent
       paint.  Several detection methods are under consideration:

I love that this was ostensibly written by Vint Cerf.

It's listed in his computer science bibliography https://dblp.org/pid/c/VintonGCerf.html and https://en.wikipedia.org/wiki/Vint_Cerf#Author

Though the edit for that authorship to the RFC came much later. https://datatracker.ietf.org/doc/rfc1217/history/

Could these not be jammed by blasting the same wavelength laser at said geostationary satellite?

Please correct me if I'm wrong, but I guess if you aim well enough, there could be a very long, narrow, non-reflective cylinder in front of the receiver that would block all light that is not coming exactly from the direction of the target satellite.

"If you aim well enough" is doing a ton of work there. Precise real-time optical tracking of a satellite from a moving platform is an extremely difficult problem. Even if the satellite itself is geostationary, it would also have to rotate to keep the "cylinder" pointed in the right direction to maintain signal.

I suppose you could make a "cylinder" or "cone" broad enough that, if the threat was static, could blot-out attempted jamming from only certain regions while staying open facing toward friendly zones.

It's a geostationary sat. It doesn't move.

No, but the airplane it would be talking to does. Hard enough when your transceiver is wide open, if you narrow your FOV to a thin cone in order to block jamming signals, the GEO now has to physically track the airplane somehow.

Either the whole satellite rotates or the transciever is on a mount that can rotate

Unless you plan on having 1 satellite per airplane, something tells me it's harder to constrain the FOV than you might suggest. There's also the small problem of the energy, complexity, & weight of having motorized parts on the satellite (or fine-grained attitude control for the satellite itself to track the craft).

Agreed, my point is it's a lot harder than tiagod made it sound.

It also doesn't account for some kind of mobile jammer making it inside the cone, particularly if it's staring at an adversarial nation where secure comms would be needed the most, but the adversary would have freedom of movement.

your assuming the target satellite doesn't reflect?

You will probably need to increase the gain (better lens, photomultipliers) on the receiver photodiode too.

Getting it to work with one end stationary first sounds like a reasonable development plan. LEO adds a lot of complexity, but with huge benefits.

OTOH the number of engineers that focus on throughput over latency is quite staggering.

I guess if your goal is just to stream aircraft telemetry and black box like recordings then latency may not be high on the agenda.

Black box data doesn't need that crazy throughput either though. Traditional RF is much easier to get right, and works even when the aircraft starts losing track of where it is and stops being able to track the satellite with its laser

I think it's the opposite? For small telemetry you want it now, but for the big data products there's no hope of "now" and so you settle for soon.

Leo seems easier to me. Geostationary is really far away. Leo is much, much closer. It's easier to hit a buck thats running right past you than to hit a stationary target across the Atlantic.

Especially if you yourself are on a moving platform

I’ll take 500ms ping for those speeds while temporarily on a plane.

No doubt! I’ve measured literal 5 minute ping times on airplanes. 300,000ms. Where are the buffering the packets!?

My guess is that you're getting retransmissions because of dropped frames, not because there's some huge buffer in the sky.

Indicated airspeed 280kts, ground speed 470kts, FL410, the packets are trying to catch up…

I like "huge buffer in the sky".

That's where I imagine all my deleted data goes.

we're all just riding the ring buffer of samsara, maaan

There’s one huge buffer in the sky!

The huge buffers are at the two endpoints (:->

Geostationary is easier to hit than a LEO constellation like Starlink. With an LEO target you need to switch at least every 2-4 minutes, Starlink ground stations can switch multiple times per minute but that's for obstacle avoidance in the air you'd only have to switch when the current target moves out of LOS entirely.

> downloading an HD film takes only seconds.

My apprehensions about living in GSO are over!

The rate that comms are colonizing orbit, or I should say the acceleration rate, feels like it can’t last for more than another decade or so before hundreds of thousands of satellites become each others Kessler nightmare.

But somehow, I expect systems will evolve to get much denser.

it’s probable not too early to consider future ring structures, for mounted satellites? Or an optimized distribution of long parking “star-bars”. As apposed to the free for all?

I expect reserved altitude shells, for space stations, depots and rocket/ship orbital maneuvers without the dodgeball.

Nice, if you want a bit more details on the TNO side https://www.tno.nl/en/newsroom/2026/02/airbus-tno-demonstrat... relying on https://connectivity.esa.int/archives/projects/ultraair

I'm really curious how the tracking works in such a system, and how "bad" the beam spread is (my impression is that from the diffraction limit alone the beam has to be spread over at least a ~10m radius after travelling 36000km).

Some info on the laser itself would also be very interesting (power? wavelength?).

Really cool project though!

> and how "bad" the beam spread is

The spread makes the tracking easier, I suppose.

Perhaps a little, however. Different paths through the atmosphere will perturb the phase of the signal; depending on conditions not all of that ~10m beam width is going to decode with an acceptable bit error rate.

Tracking and actuation is nothing new or particularly challenging, IMHO. It's the laser/optical part combined with throughput at that distance that is the main area of R&D, I think.

> These developments entail a future where travellers could enjoy reliable, high‑speed internet while flying, and where people on ships or in vehicles crossing remote regions can stay connected without interruption.

How reliable/feasible would this be on the ground? From what I understand, shining non-trivial lasers in the sky is a massive liability because of the potential to interfere with aircraft. I don't see anything about the wavelength used, but even if it's outside the visible spectrum, it would still be subject to interference from aircraft when used on the ground or at sea.

Optical links are being developed for use from fixed ground stations.

https://news.ycombinator.com/item?id=46709548 - Discussion from a month ago with several links for a recent example.

It's being implemented. I thought I saw that Amazon Leo (nee Kuiper) was going to lean on it pretty heavily.

https://www.techbriefs.com/component/content/article/47300-u...

That talks about inter-satellite links (which Starlink uses already). Parent comment asked about ground <-> sat

Some miniaturization required.

But that means you need to have a different laser pointed at every single individual aircraft right? Doesn’t really scale.

I suppose you can do time-sharing. And use mems-mirrors to quickly move the beam between different targets.

Laser TDMA! :-)

https://en.wikipedia.org/wiki/Time-division_multiple_access

You can probably do phased arrays. (It might already be a phased array.)

Pretty sure phased array LASERs are not yet a thing.

I was not sure, but they are!

https://cga.anu.edu.au/research/activities/laser-beam-steeri...

https://www.darpa.mil/research/programs/excalibur

I guess in some ways even the fancy multi diode fiber lasers are phased arrays, just with the single goal of higher output power.

Looks like these are in early development and nowhere near ready like this test was.

Lasers are coherent emitters; you can definitely make interference patterns with them, so I don't see why LASER MIMO wouldn't be possible, in theory.

Yeah but this is research, if they're to come up somewhere, where else would it be?

If starlink satellites get laser downlink, it might work :P

laser downlink to one point, isn't it? Not to 300 moving aircrafts at once.

How does Air force one accomplish their data connection?

Air Force One (and all of the other US flying command posts) are basically giant collections of various antennas.

Here's an article from 2017 about (then) recent installation of what were almost certainly satellite communication antenna.

https://www.twz.com/10470/air-force-one-jet-reemerges-with-u...

Impressive! I believe round trip latency would be 0.5 seconds.

That's ~162.5 MB in transit at any time

Excellent for pingfs (https://github.com/yarrick/pingfs)

There's a patent (2017/0280211 A1) for using this as a data storage method, and there was a company called Lyteloop trying to leverage the idea for data storage with estimations for petabytes across constellation.

That could you used like RAM like the delay-line memory used by early computers!

Shouldn't it be 1000/16 = 62.5? Impressive nonetheless, of course!

The article says 2.6 gigabits/second which is 2,600,000,000 bits/second, 2,600,000,000b/s * 0.5s / 8 is 162,500,000 bytes, 162,500,000 / 1,000,000 is 162.5 megabytes

Right, thanks

Weird.

I marvel at the ability to track a target in both directions ~40k+ km away while moving quickly (kinematic) considering atmospheric and relativistic effects.

Kinda cool but seems paltry during a time when we have Starlink, which also has space-to-space laser links between satellites, of 100-200Gbps in bandwidth.

Though I suppose this is a bit safer from Kessler syndrome

[dead]

> Because laser beams spread far less than radio waves, they provide more secure links

Basing your security on laser diffusion seems sus.

It's worth it as another layer of security. The beam width being so narrow means even intercepting it becomes harder. This is more relevant for down-to-earth links where the spot hitting the earth is so narrow it could be confined withing a geographically controlled area, rather than hitting an entire continent like longer wavelengths do.

These beams are much harder to detect and eavesdrop upon. You increase the difficulty for a remote attacker. I wouldn't stop encrypting the data, however: The Alphasat TDP‑1 has a telescope with an 135mm aperture. The beam diameter is likely to be at least 700m wide according to the diffraction limit.

All security is based on a combination of individually flimsy ideas