SpaceX's Starlink Is Testing Internet Service for Aircraft (bloomberg.com) 46
SpaceX's Starlink unit is testing its space-based internet service with several aircraft and wants to offer in-flight connection to airlines "as soon as possible," a company vice president said. From a report: Starlink is in talks with several airlines about offering in-flight broadband connections, Jonathan Hofeller, vice president of commercial sales, said Tuesday on a panel at the Airline Passenger Experience Association gathering in Long Beach, California. That would put the company in direct competition with Viasat, Intelsat SA, Telesat and others. Starlink is producing six satellites a week at its assembly site near Seattle, Hofeller said. It is also moving to a more sophisticated version.
Internet for Aircraft? (Score:3)
Re: (Score:3, Informative)
Re: (Score:2)
i got my beta invite last february; signed up and put down my deposit literally the same minute -- and still have not received anything.
I'm guessing that their having trouble producing the hardware to meet demand; but that's just a guess.
Re: (Score:2)
I'm guessing that their having trouble producing the hardware to meet demand
they're. dammit
Re: (Score:3)
To provide service, the Starlink satellites require an endpoint on the surface. If such an endpoint is not available they will not be able to provide service. They were investigating the potential of relaying the signal between satellites before going to the surface but I do not think the current satellites support such communications.
One other possibility is that the satellites do not orbit far enough North / South to provide you with service. If you look at how the satellites orbit, you will notice
Re: (Score:3)
Read the article. Shortage of semiconductors is one reason for the delay. As others have commented, the "not yet available" capability of satellite-to-satellite relay (via laser) somewhat limits who can receive service. Capacity limits is another (only so many users can be served within a specific geographic area, think "miles", not "hundreds"
Re: (Score:2)
Thanks for the link, it is great. Unfortunately, my home town is a bit to far north.
What is interesting is that there are endpoints in Alaska but no satellite coverage. There appears to be a new constellation with an orbit that travels over Alaska, but it is currently moving into position. Guess we will have to wait to see where it ends up. One assumes that global coverage is an eventual goal.
Re: (Score:2)
I'd be a little surprised if they ever got the laser link to work reliably. It requires a narrow beam to hit a small moving target, from a moving platform, from an extreme distance away.
The northern coverage will be essential for many air routes that go that way. The quickest way between two points is a straight line, and that often involves going north when travelling from say Europe to the Far East, or to a lesser extent Europe to the US.
Re: (Score:2)
I'd be a little surprised if they ever got the laser link to work reliably. It requires a narrow beam to hit a small moving target, from a moving platform, from an extreme distance away.
In November 2001, the world's first laser intersatellite link was achieved in space by the European Space Agency satellite Artemis, providing an optical data transmission link with the CNES Earth observation satellite SPOT 4. It achieved 50 Mbps across 40,000 km, the distance between LEO and GEO. In 2014, ESA launched an upgraded system capable of 1.8 gigabits per second over the same distance which is still in operation today. JAXA first demonstrated Earth to LEO laser communication in 1995, at 1 megabi
Re: (Score:2)
Thanks, that's interesting. I guess the next issue will be bandwidth. All the ones demonstrated seem to be sub 1gbps.
Re: (Score:2)
Thanks, that's interesting. I guess the next issue will be bandwidth. All the ones demonstrated seem to be sub 1gbps.
Except for the one ESA is using between LEO and GEO, which is 1.8 gbps.
Re: (Score:2)
I'm in the PNW; there's a house just down the road from us (2 or 3 miles?) that has their dish. Also oddly (and annoyingly) enough guy i worked with who lives in town was able to get starlink to replace comcast. But, hard to say.. either way, starlink has the potential to absolutely obliterate traditional sat providers.
That said though, we're currently using a pair of tmobile hotspots; luckily we have LOS (though at an extreme distance) to some cell towers; it's passable, just kind of expensive. ~$140 for
Re: (Score:2)
Re: (Score:2)
pretty much this; unless you've had to use viasat or similar geosynchronous service you don't understand just how bad the service really is.
minimum round trip ping time over viasat is ~600ms, that's under the best of conditions. it was usually more; and then during 'peak' hours (read as: 5pm til about 10pm) it was substantially worse.
4g hotspot i'm on now is clocking 40-50ms.
and i believe starlink is in the mid-teens.
and that's just latency. throughput is a similar tale of woe.
Re:Internet for Aircraft? (Score:5, Interesting)
Not sure where you are based, but both in Europe and the US it's already available for some time. Here is a live interactive map of the coverage: https://satellitemap.space/ [satellitemap.space]
What makes you think they are having problems, and what problems might that be?
Ohio here. Joined the beta in January '21, ETA was "late 2021"
Just last week got an update email, ETA is now "Mid to end of 2022"
Go look at the coverage map you linked. Ohio is 80% within one cell with no ground stations. Another 10% in another cell to the east with no ground stations.
Only a 10% slice to the north shared with PA has one ground station, which isn't in view of birds over 90% of the state.
This was the same back in January as it is today.
Now zoom out and look at all the cells without any ground station within them.
Something like 60-70% of the USA is *NOT* covered.
I'll assume you're equally incorrect about Europe and less than half coverage is there too.
That's a single digit percentage of the globe where starlink is available.
That does not "already available" make.
Once starlink gets their secondary orbit satellites up, where the low orbit satellites will be able to "hop" communications between them to reach further away ground stations, service will technically be "available" but at significantly higher latency than ideal.
There's been no updates on when that will happen either.
Re: (Score:2)
What about it?
I am sure that they can sell wireless internet services to multiple groups at once.
Re: (Score:1)
Its viability has always been in question. The claim somewhere to keep 30K satellites up and running and replaced every 3 to 5 years you have to have a stream of $30Billion per year. So at $100 per month clients you have to have 25 million customers. /maintenance.
Do you think there are 25 million people in the world who can afford $100 per month that have no access to fibre or 5G. non rural areas in the world including third world countries are being fibred or are going to have 5G towers at much lower cost
Re:Internet for Aircraft? (Score:5, Informative)
The other place $30B appears is "Musk said Starlink could make up to $30 billion in revenue per year."
So, if the high-end cost totals $30B, and the high-end revenue is $30B per year, you're sitting pretty unless the system costs as much to sustain every year as it did to initially construct, which makes no sense.
Anyways $30 B is a high-end - not most likely - estimate for both buildout and annual revenue. The system is scalable - it doesn't need anywhere near 30K satellites to provide global coverage - so they can stop scaling it when diminishing returns kick in.
All of which is not to say it's safe and not risky financial bet. But it's not true they expect to just break even or anything like that.
https://www.morningbrew.com/da... [morningbrew.com]
Re: (Score:3)
Musk estimates need to be taken with a few kilos of salt.
The interesting part is how they will mount a receiver on an aircraft. The current one isn't exactly low drag. It will need to be custom fitted to every model of aircraft, and certified to make sure it's safe in all conditions.
Discussions with airlines are probably "get back to us when you have got it certified for our aircraft".
Re: Internet for Aircraft? (Score:3)
Re: (Score:2)
It would be nice to think that, but experience suggests that Musk's tweets cannot be relied up and are rarely reviewed. This kind of thing is what the SEC was talking about.
Re: (Score:2)
Nah. It's in the MBA 101 playbook.
"Musk wrote in the email that he planned to take the long Thanksgiving holiday off. But, after discovering the Raptor situation, Musk said he would personally work on the engine production line through Friday night and into the weekend."
How else do you get your workforce to "voluntarily" give up their holiday vacations?
Re: (Score:2)
Presumably this would replace existing satellite dishes currently used for satellite internet on aircraft. The standard starlink dish is smaller than e.g. viasat's (about 23 vs 30 inch) so it should easily fit in the same radomes.
Re: (Score:2)
The interesting part is how they will mount a receiver on an aircraft. The current one isn't exactly low drag. It will need to be custom fitted to every model of aircraft, and certified to make sure it's safe in all conditions.
There are at least two branches of the U.S. military that are developing Starlink communications technoilogy for aircraft. I have seen at least one press release about this (U.S.A.F) and have a relative that described Navy doing the same.
That doesn't immediately make it available to commercial aviation but it does imply it's not a whacked out idea. It can be done. But no, it's not Dishy McFlatface mounted to a fuselage. It's much more sophisticated technology, but as usual, most of it trickles down
Re:Internet for Aircraft? (Score:4, Insightful)
"Do you think there are 25 million people in the world who can afford $100 per month "
There are 50.000 ships running the oceans and the logistics firms would love to have real-time feeds on every sensor (temperature, shock etc) in every single container, with 24000 containers on one ship that alone should pay for it.
Not to mention that the governments would love to have live feeds from fishing boats, and thousands of ocean sensor stations, whale programs and whatnot.
Marine drones possibilities alone are endless, ditto for drones in the arctic and antarctic.
Every penguin and polar bear gets a Bill Gates chip under the skin, you'll see. :-)
Re: (Score:2)
"And they already get that cheaper using traditional satellite internet."
3 satellites for 25 million?
I'd prefer 30-40.000.
Re: (Score:2)
Shipping companies won't want live feeds of telemetry from containers. They only care about meeting certain requirements and making insurance claims, so a simple logging sensor that can be read back at the destination is all that is required.
A simple sensor can easily run on a small battery. A connected one that relays data in realtime needs a much larger battery, or power connection to the connex. A ship powered sensor could easily be defeated by simply unplugging the power supply too.
The IoT and safety st
Re: (Score:1)
The service is in beta, it's hardly been rolled out in a pre-release, or had any promises made. Up until a few weeks ago, it was advertised as "Better Than Nothing."
The limiting factor is phased-array dish production. Obviously the antenna for an aircraft is going to be entirely different and a bespoke product, and each one will service dramatically more individuals.
This has always been part of the plan for Starlink, along with mobile backhaul, and military applications.
Re: (Score:2)
"Makes me wonder if their having problems and are looking for ways to salvage things."
They are building antennas to sell as fast as they possibly can.
Re: (Score:3)
Aircraft are much more profitable then. On a commercial flight outside the US, you can pay $20-50 for internet for the flight. Some flights even charge by the megabyte (one flight I was on gave everyone in business class 200MB of internet for free).
The GA market for internet is even richer - it's right now only available on
Handoff frequency (Score:4, Interesting)
Now that Starlink has LEO satellites that are going fast enough with a big enough footprint, I'm not surprised the average airplane is nothing more than a (very large) blip in their coverage.
I suspect marketing looks at this large blip and salivates, but engineering and management don't even notice.
Re: (Score:2)
Back in the 00's I regularly used my cell phone on flights and was able to use terrestrial cell phone towers
The 10 mile range on land is due to the curvature of the Earth and horizon, you have much longer reception from the elevation of a commercial flight
I have not tried recently, what with 911 legislation and chance of being called out as a terrorist ;)
imo, the airlines just want to be able to charge exorbitant prices for internet access and all the scare stories about brining a plane down with your cell
They can't provide the coverage (Score:4, Interesting)
The real situation is that it is clear their current first shell of satellites can not actually provide coverage to most of the US. It can hit a few random cells here and there, and what they have done is prioritize the cells with the most pre-orders to jack up their subscriber numbers. Meanwhile, the people in rural US areas are being skipped over in preference to these more populated cells, unless you happen to be adjacent to one of the ones they picked. The chip shortage is to provide them cover to not have to admit their satellites can't provide this complete coverage.
There are many stories of people in areas who pre-ordered the very first day (myself included) who live in very rural areas who just got delayed. I am within range of a ground station, and am in an area that definitely could be covered (there is coverage not that far away), but they are skipping my cell and others around me because it is clear their satellites can't handle it.
So I don't see how they would handle aircraft coverage until they can cover all cells, as planes fly over me all the time. It would be a horrible patchwork of coverage as these planes flew in and out of service areas and completely unusable. It is the same reason that them saying they are going to have mobile coverage soon is a complete fabrication. You can't do that until you have all cells in the lower US open, or at least almost all of them. Right now, it is not even close and my guess is Elon found out he can't actually cover the landmass with the current satellites and only gets spotty coverage.
So if you think you are getting mobile or aircraft any time soon, just check in on your Tesla CyberTruck order at the same time and see how that one is going as well. They used pre-orders to basically take a hundred bucks from everyone to figure out which cells they should turn on, and which ones to leave off until the next batch of satellites someday make it to orbit....
Re: (Score:2)
The real situation is that it is clear their current first shell of satellites can not actually provide coverage to most of the US. It can hit a few random cells here and there, and what they have done is prioritize the cells with the most pre-orders to jack up their subscriber numbers. Meanwhile, the people in rural US areas are being skipped over in preference to these more populated cells, unless you happen to be adjacent to one of the ones they picked.
The current satellites provide coverage of the entire continental US. What's missing is receiving ground stations in each cell and user terminals. They've told reporters they're making 5000 user terminals per week right now. They had 600,000 preorders. More than 100,000 have been fulfilled so far. At their current production rate, it will take almost two years just to build all the additional user terminals they need. Obviously they're trying to ramp up, but just as obviously, the supply chain issues
Re: (Score:2)
Re: (Score:2)
You can go online to an active cell today, place an order, and get a ground terminal within a few weeks.
You keep saying this. Got a citation better than a Reddit thread? SpaceX themselves aren't talking. I'm willing to believe they're playing a little fast and loose with the pre-order queue, but 5000 per week really is a drop in the bucket for a queue as large as theirs. I think we can safely assume they quoted their best week, too, and most of the past year production was lower than that.
The active cells are patchwork all over the US, and it has nothing to do with ground stations to relay, as many spots are within range of ground stations.
Are they? Beam forming to ground station relays are just as subject to limitations as beam forming to user terminals.
Finally... (Score:2)
...in air internet that won't suck... as much.