But it seems like a recipe that combines so many of the worst parts of running a groundstation. The setup can't easily be specialized enough to really milk capability (say, different ways to correct pointing and orbit determination errors that need more direct control of the antenna than just feeding a TLE), but it's likely so general that there's still significant engineering effort for anybody who wants to use it. If you want to do anything requiring round-trips to the satellite, where you need to ship raw radio signals to a DC, process them in software to demodulate and decode, then encode and modulate your responce before shipping it back as raw radio signals to the groundstation, sorry but that latency is painful and hurts you on link utilization.
In comparison, buying a small turn-key dish and operating it is not that bad. Paying per-minute or per-pass gets really expensive for smallsat/cubesat operators, at least at the usual prices I've seen and compared to the operating cost of equipment you own. Also, if you're just running a tech demo then it's not exactly prohibitive to partner with somebody who has excess capacity.
I think this offering is missing the mark. If I had to make an analogy, I'd say they're offering roughly "GCP for downlink" to a world that's not quite homogeneous enough for it to make sense. If I were trying to do this, I'd be aiming to be the "Squarespace of downlink" (more tightly scoped capabilities, but much better performance and more turn-key) or someone delivering groundstation-in-a-box kits.
Context: I used to work at Planet Labs, having spent considerable time collaborating with the groundstations team, as well as some collaboration with the missions ops team.
First of all, could you give some concrete examples of the kinds of data people using this would be handling? The examples provided in the article are incredibly vague, and the linked Wikipedia articles didn't help clarify.
Could you give some ballpark numbers and breakdown of the costs involved? How much does it cost to manufacture and launch a small satellite? What about buying a dish and operating your own ground station? What about regulatory fees for licensing and registration? There's a few mentions of hobbyists, but I have a hard time imagining many people with enough money and interest to deal with all these challenges. Is there some secret cabal of millionaires and billionaires that are really into space?
Planet operates imagery satellites, and as far as I know that's the bulk of commercial data _produced_ in orbit that needs to be downlinked. Other commercial services that
could include radar data (ICEYE, Capella), ship/plane monitoring (Spire, Iridium NEXT), or high-latency low-bandwidth data (think devices out in the field, on vehicles, with relatively light telemetry and/or control needs). Telecom is usually "bent pipe" (think dumb repeaters) and not any kind of store and forward setup that would use this.
In the future, I could see some broadening of the field with more tech demo/development that needs dedicated hardware, and the possibility of commercial experiments run in an automated orbiting lab downlinking data.
Of course, all of those also need what's called TT&C (Telemetry, Tracking, and Control) links, which I've nearly always seen on a separate physical radio. That tends to be a less demanding link, and more likely to be able to make use of something like this IMO.
> Could you give some ballpark numbers and breakdown of the costs involved?
Unfortunately, I very much cannot do more than point at what's already public; that's one thing Planet is very tight-lipped about. The idea that one spends about as much on launch as on the hardware they launch seems about right, you can go see what Rocketlab and SpaceX quote as prices. I'm sure you can imagine there's still a lot of negotiations once you're talking about big stickers.
Someone's already mentioned SatNOGS further down in the comments. They seem neat although I haven't looked too much into them yet. Schools are spinning up a surprising number of satellite projects, and amateurs are already developing hardware, running tests, and practicing with radioing the hardware that's already up there. I'm sure it's not far from some pitching together on the group buy of a very small amount of launch space.
> Is there some secret cabal of millionaires and billionaires that are really into space?
Yes, but they're not so secret. Elon Musk, Jeff Bezos, Richard Branson, off the top of my head. DFJ and Data Collective are VCs that have been fairly interested, I'm sure there are more looking to get a piece now.
also, ditto what he said. This news was certainly an interesting conversation that happened at Planet HQ today, but a couple of folks knew the people doing this work in Amazon before it was announced today. There are a lot of regulatory, licensing, and construction hurdles. Not the mention, the placement of the ground stations is going to be crucial, but otherwise...I can see upstarts in the smallsat industry totally taking advantage of this should their orbits compliment the deployment sites of the ground stations.
We build and operate our own ground stations for the most part and I can understand/sympathize that we're fortunate enough to do so since much of our business depends on it. It's a costly exercise for sure...
"Give us $10B to build your datacenter, and we'll throw in a ground station for free!"
They say it's going to save at least 80% over the cost of operating your own, can only assume they will price it very competitively and have done the math. Hard to say anything without seeing numbers.
I've seen new turn-key installations come with an equivalent UI and an API to match it. Plus a bunch of debugging/calibration/testing controls. And then (assuming licensing, real estate, and services availability) you get to drop that "anywhere" and optimize its geographical location to match up with your system requirements.
Turn-key systems will totally let you specialize it to maximize utility. Selection of RF feeds/hardware on the input side, and eg. hardware data modems and co-located command-and-control servers on the output side.
That said, there was significant difficulty in getting a dedicated groundstation due to real estate and licensing issues. So I think that it is still an attractive option for those who may not be able to furnish their own ground station depending on the constraints of the mission.
Of course, I'm biased; I say this as a generalist who has done work inside the equipment room at the base of dishes, hooking up my own hardware and managing live passes, as well as debugging via interactive sessions with satellites.
Please explain this, I don't understand why it would be the case.
Likewise, people that are concerned about optimising the radio link are not likely to accept restrictions in the receiver design for instance.
So I'm curious what factors you are thinking of that would not make sense as part of the service offering but would be needed by would-be customers of the service?
Here's the thing though. There's very little matter even in low earth orbit, so you can't use matter to transfer heat, as you would do on Earth (conduction and convection are out). That leaves only radiation, which requires a pretty large surface area. Try placing a computer on vacuum even on Earth and report back how the temperature looks like.
So no. Getting rid of heat is a big issue over there. Even in the shade.
But more importantly, space is empty, and the best way to cool is to dump heat in some medium which carries it away (air is the basic one on Earth, but applications that really need cooling like to use water as a primary medium.) In space you've got...nothing, basically, somyou are stuck with radiating, which works, but poorly.
On Earth, air molecules can carry heat away from you. You transfer heat to the air around you. In space, there is almost no matter around you to absorb the heat and carry it away from you.
Your only option is to radiate heat away from you in the form of infrared light but that is a slow process.
So ... the heat stays put and doesn’t go anywhere?
Also, if you think about this in the context of the Sun, you get a really visceral feel for just how much energy is being produced. Black-body alone is transferring that energy to us. Scary.
Nope, hence leveraging Amazon et al. I would consider the AWS "above the cloud" offering (as I've now dubbed it) to essentially be the same commoditized thing as what you linked. An extraterrestrial CDN with edge compute if you will.
All of this equipment can usually fit within a rack or two. The downlink stuff can fit within a few U. Uplink involves some big power amplifiers. Back in the 60s it would have involved dozens of racks of equipment.
"Underwater", meanwhile, is more of a mode of operation / design for data centers, with the goal of that simplifying (and decreasing costs for) cooling.
An underwater data center doesn't imply that the data center is actually off-shore to any large degree. I'd imagine they'd look more like undersea-cable landing points, built "into" a beach.
If a published study on this exists, I'd love to take a look.
It'll certainly not be worse for the environment than any normal data centre, unless it springs a leak and something toxic escapes.
Then again, I'm a bit biased - I worked on both Phase 1 and 2.
Also, I guess it would solve the problem of it being expensive to put things in space by virtue of somehow having made someone else pay for vast overcapacity of sensors.
perennial fear of every small company dealing with much larger partners.
Amazon's offering is geared more towards ground stations, but they might move up the stack and start providing data refinery-type services on top of the ground station work.
Oh, our entire stack is built on Google Cloud Platform.
Also, what's the TAM in your specific market vs the AWS product's market? How has the TAM changed in the past 5 years?
Lastly, I heard your head of engineering brews better beer than any of your competitors. Is that true? Can be provide samples?
Examples include Land-Cover-Mapping (mapping pixels to classes like forests, urban areas, water, etc.) which can then further be used to do crop monitoring or land-use monitoring.
I guess this is different than the product AWS is offering here, which is more about getting the data from/to the satellite, but not about processing (at least for now).
Satellite data imo sucks, especially aerial imagery. Too many damn clouds to get anything useful in real time haha!
Labs looks very interesting, very lofty yet achievable goals.
I wonder if the same principle could work for telescopes, or are radio telescope designs far too specialized to be a commodity? Huge installations funded in spite political boundaries might be possible in such a scenario
Aka reaching beyond the cloud.
How much demand would there be for this type of offering?
Usually the export laws and restrictions (ITAR) refer to the technology used to launch the satellites, not the data they downlink.
But if you had the key, you could also use encrypted satellite data.
(Edit note: the parent changed their text to say intelligence data, which is obviously not what I was talking about)
Encryption is still not necessarily required, unless you are landing the signal only on a groundstation in the United States.
AWS can handle ITAR controlled data though. They already have the AWS GovCloud for data subjet to ITAR restrictions.
But usually the science type or payload data is one thing, and then the lower level hardware telemetry is done in a different way.
I've used GovCloud to store ITAR data. It's cool. If you encrypt your ITAR data, you can also store it in a public cloud like S3, but just for storage, you shouldn't decrypt it there or have the keys there.
Source: I worked on the telemetry team at SpaceX.
As a constructive counter example, there are plenty of free services you can tune into.
Literally in at least one case. The NRA donated two surveillance satellites to NASA back in 2012  and the plan is to fly one of them as the WFIRST telescope . I don't think there is a launch date yet.
Some telemetry and sensor data may be EAR99, which is as far as export restrictions go is about as nonrestrictive as it gets. For example iTunes is EAR99— that's why the EULA says you're not allowed to use it to develop nuclear or biological weapons.
I wonder if this would work with Blockstream Satellite: https://blockstream.com/satellite/
(Of course using a remote ground station operated by someone else completely defeats the purpose of Blockstream Satellite, but it would be a fun "hello world" for AWS Ground Station, if it were self-serve)
The SATNOGS receivers, usually a low-cost RTLSDR, cover a much wider spectrum than just amateur bands (which prohibits commercial use of it's spectrum), so I wouldn't be surprised if non-amateur entities are eyeing the service, or if SATNOGS organization is thinking about going commerical (although there's already competition in this field besides AWS Ground Station).
I usually say it's 12-18 months between when an AWS service is announced and when it's actually useful and stable enough to use in production. In this case, I'd easily double that.
(Also, the "imaginary" sat that was added to the account in this case was in fact NOAA 15. That one actually transmits realtime imagery to the ground that anyone can receive).
It’s far cheaper to send bits to space than actual matter on a rocket.
It would also promote re-use of satellites and mitigate Kessler syndrome (unless induced demand makes it worse).
Also, I'm guessing you need to provide your NORAD ID and FCC license so these dishes don't get used for sigint.
plz fix you current shit AWS instead of introducing new shit every week
Of course, setting up a radio receiver on the ground is not the hard part about launching satellites, so this is unlikely to change much. When AWS starts letting you launch satellites for $1.99/pound, then we can start worrying.
Also, I think they kind of jumped the shark with this one.