Slashdot Mirror
FCC Grants OneWeb Approval To Launch Over 700 Satellites For 'Space Internet' (theverge.com)
OneWeb has been granted approval from the FCC to launch a network of internet-beaming satellites into orbit. FCC chairman Ajit Pai said in a statement: "Humans have long sought inspiration from the stars, from the ancient Egyptians orienting the
pyramids toward certain stars to the Greeks using constellations to write their mythology. In modern
times, we've done the same, with over 1,000 active satellites currently in orbit. Today, the FCC harnesses
that inspiration as we seek to make the promise of high-speed internet access a reality for more Americans, partly through the skies..." The Verge reports: OneWeb plans to launch a constellation of 720 low-Earth orbit satellites using non-geostationary satellite orbit (NGSO) technology in order to provide global, high-speed broadband. The company's goal has far-reaching implications, and would provide internet to rural and hard-to-reach areas that currently have little access to internet connectivity. Additionally, OneWeb has a targets of "connecting every unconnected school" by 2022, and "bridging the digital divide" by 2027. According to OneWeb, the company plans to launch an initial 10 production satellites in early 2018, which, pending tests, will then be followed by a full launch as early as 2019.
36,000 kilometers for geosynchronous orbit, versus 200ish km for a low earth orbit.
36000 kilometers is 0.120 seconds at the speed of light, there and back is your 250 ms. Low earth orbit could be much faster.
What I'm wondering though isn't this something nasa/faa should be approving? That's a lot of potential space trash.
at 700 satellites, economies of scale come into play, so cost per additional satellite is probably a lot cheaper than $50 million or whatever.
Iridium satellites cost about $5 million per piece. So your cost estimate is about 10 times to high. Launching and operating will cost a pretty penny too but if the system supports about 10 million subscribers the cost of a subscription will be in the same order as a dsl or cable subscription.
The FCC approved use of the US frequency bands.
The satellite launches themselves area separate issue. They might try and keep the costs down and launch outside the US.
Its communications to the world. Imagine owners of a nations city telco networks and regional cell spectrum looking up and seeing real competition for the first time.
People who need voice, the cloud, data all the time in very isolated areas. A device might have to log data all day until its finally connected to a network later.
Now services can communicate in real time, all day.
Very restricted and expensive remote telco services will face competition. The days of changing a lot for speed and not much data could finally change.
Networks that have bad quality or no voice solutions due to not been able to work on very isolated voice services.
Domestic spying is now "Benign Information Gathering"
The cost of the Iridium project went well into the billions (I've seen $4B) for 77 satellites. The $50 million per satellite figure is not too far off. The $5 million per satellite probably only covers only the manufacturing and not the tooling, R&D, logistics, etc..., anyways, it is obviously not representative of reality.
Iridium NEXT has a budget of $3 billion, a little cheaper but still in the same ballpark.
I am always skeptical when people announce huge savings. First, it isn't like private companies love spending money for nothing, if there is a way to make satellites 10 times cheaper, it will definitely be considered. Usually, the reality is that when some part of the process becomes highly optimized, some other part become the bottleneck. For example, let's imagine there is a way to make satellites for free, the launch, maintenance, ground stations, ... will become significant and it may turn out the savings won't be quite as much as expected.
Wasn't SpaceX planning to do something similar? But with more than 7500 satellites?
I think a major problem with making cheap satellites is that you need specialized ("space rated") parts, and that vendors of these parts will charge inflated prices, just because they can.
This is a problem that SpaceX had with their attempts to make cheap rockets. Their solution was to develop a lot of things in-house. They also buy parts from other vendors, but they make it clear they want a fair price, otherwise they'll walk away and find another solution.
Should we be worried about the "Kessler Syndrome"? That's where the density of objects in a given orbital volume gets to the point where a single collision causes a large amount of debris which in turn causes more collisions which ...
The 700 new objects will be put into LEO where, in order to provide worldwide coverage they won't be in a single orbital plane (like the "Clarke belt" or geosynchronous orbit). Instead they, like GPS or Iridium will be crisscrossing with each other (no problem if properly designed) but more importantly with all the other "junk" there (like the space station!). In this most densely populated volume of space, encounters of the worst kind can/may/will happen.
Does anyone know how much this closer to the tipping point this will bring us? Is this system's effects negligible in comparison to the clouds of debris from various A-sat tests? What about Elon Musk's proposed system (which I think also got approval) which had several thousand (four thousand?) similarly LEO satellites?
I hope this doesn't come across as either critical or flaime-bait.
Having always been fascinated by space, I'm always keenly interested in any launches. The SpaceX approach to media, with live-streamed launches, has been mesmerising. But it occurs to me that, as a planet/species, we're now putting more and more into space than at any time since the launch of Sputnik. Of course, different countries have different governmental controls put in place to license companies for aerospace operations. This is entirely sensible, since a mis-fired rocket could easily cause an incident with an aircraft, or land near a populated area, or worse.
But at what point do we realise that we can't simply have endless, uncontrolled launches into space; that perhaps we need to have some form of [perhaps UN-backed] international framework to ensure that there is full coordination and collaboration on our use of local space, orbits and launch windows.
Or did that happen and I just didn't get the memo?
The Giza-Orion thing aside, what most Egyptologists seem to agree on is that they used the stars to fix the North-South orientation of all their pyramids. (Because that orientation drifted over the years they stared building them, following the precession of the earth axis)
I'm skeptical of the figures too, but I'll wait and see the math. LEO is a lot cheaper to reach than GEO, and you are well within the Van Allen belts so don't need as much radiation hardening and associated mass, both of which are going to bring the price per launch down substantially. Of course, the plan also involves nearly 10x as many satellites, albeit presumably much smaller and more "disposable", which will push it back up again.
From the illustrations on OneWeb's website it appears that we're essentially talking about a few hundred slightly oversized cube-sats that could potentially be thrown up a few dozen at a time by SpaceX's Falcon 9 heavy or a similar booster, so you could easily end up with a smaller price tag than Iridium. Still likely to have a total price tag of a few billion, but not tens of billions, and potentially still commercially viable if you can resell enough bandwidth at the low, low prices that are all that their primary customers can afford. It's all going to depend on the unit cost and how many they can launch per booster - if they can bring both of those down low enough, provide enough bandwidth, and some higher end services (real time global tracking of ships and aircraft, perhaps?) then I don't see why it wouldn't be viable.
UNIX? They're not even circumcised! Savages!
They intend on using 18 orbital planes at an altitude of approximately 1200 km (750 miles). Doing the math for a 3000 mile round trip at the speed of light gives me 16 milliseconds. Of course, the actual latency will be higher since that 16 ms latency is just the trip to and from the satellites. You also need to add in the distance between both the satellite you connect to and the one that the ground station connects to. Worse case would be the ground station being on the opposite side of the world, in which case the total round trip latency from the user to the ground station would be 96 ms.So in summary, depending upon the relative locations of the user and the ground station the user connects to, the latency added by the satellites is between 16 and 96 milliseconds.
Oneweb is not using the "individually hand crafted" model for the their satellites, they're using a lower standard of "medical grade equipment". This makes sense given the large number of satellites they intend on using. The cost per satellite (not counting the launch costs) is estimated to be about $500,000 each. So the cost for 720 of them is $360,000,000. Of course, the actual cost will be higher since all those satellites need to launched into orbit.
It's Astrolink all over again.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
But if the government spent the money directly (on hiring more teachers, buying books, fixing leaky roofs) that would be communism which leads to hospital death panels, mandatory gay marriage, and banning SUVs.
If they hire a big contractor to build a humongous boondoggle then it's private enterprise, which is freedom and apple pie and NUMBER ONE!!!!
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Minor correction. I forgot to include the return path from the ground station. So the latency, depending upon relative locations of the user and ground station, is 16 to 175 milliseconds.
And we are pretty sure they were built by undocumented Hispanic immigrants.
Wait a minute, since when does FCC grant an approval to launch anything? I thought this was FAA's jurisdiction?
Ezekiel 23:20
Where did you get the 3000 mile figure? That seems unnecessarily high. Equidistant planes should be separated by around 1500 km at US latitudes. Together with a 1200 km altitude, the nearest orbit should be between 1200 and 1400 km away from you. Forth and back, it's between 2400 and 2800 km, not your almost 5000 km.
Ezekiel 23:20
SPAAAAAACE!
Several per launch for certain since so many are going to be in very similar orbits and probably won't be especially heavy. That kind of turns using prior examples into a very wild guess.
Good project or bad, who knows, like many other things in technology companies at the time they were fucked over by a bunch of bankers and never had a chance.
Also are you sure it wasn't going to be geostationary? With four satellites making it functional I can't really see it being anything else (but I'm no expert).
Ajit Pai said in a statement: "Humans have long sought inspiration from the stars, from the ancient Egyptians orienting the pyramids toward certain stars to the Greeks using constellations to write their mythology"
Christ, how's that for purple prose! Should a regulatory body be using language like that? Surely they should just keep to the facts of the matter.
OneWeb claims 30 ms latency, which doesn't sound too bad. My round trip time to slashdot.org is 110 ms, and it's perfectly usable.
Since when does the FAA regulate space?
Seven puppies were harmed during the making of this post.
Unless they're going to be playing Unreal World Of Steel Theft Craft 4 why is latency an issue? It's for schools, isn't it?.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Surely they should just keep to the facts of the matter.
I suspect talking this way shows a certain contempt for the intellect of others.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
Orbit is 750 miles up. Sending a packet to and from a satellite takes a round trip of 1500 miles... And then you need to receive a response from whomever you sent the initial packet to. Another 1500 miles. Total round trip distance? 3000 miles or 16ms overhead. Of course, that doesn't include any lateral distance to the actual ground station you're using. Although "ground station" may be a bit of an over statement. I could easily see a modified "user terminal" connected to a ground based high speed internet. Also, more ground stations is of benefit both to the users and to the company. Users benefit because of lower latency. The company benefits because less of the available satellite bandwidth is being used on inter-satellite cross communication routing user packets to/from the ground station.
Good enough for voice services.
Domestic spying is now "Benign Information Gathering"
Actually, they're also thinking of homes. And lower latency is better than higher. Their use cases include aircraft (business, commercial, and military), health centers, schools, libraries, and homes. Their intent is to provide internet access anywhere in the world and allow for the use of any application that uses the internet. And there are applications where low latency is required. Otherwise they could simply launch 4 satellites in a Draim constellation and be done with it.
The antennas used by the users will be a small phased array (approx 36 cm by 16 cm).
...[I]n Egypt, the Pharaohs
Had to import Hebrew braceros.
Didn't you learn that song in school?
Il n'y a pas de Planet B.
I'd be interested in knowing where you found that 30ms figure. But it does imply that Oneweb is planning a ground station approximately every 1500 miles. That in turn implies about 32 planned ground stations.
36000 kilometers is 0.120 seconds at the speed of light, there and back is your 250 ms.
If you're talking round-trip ping times to a server for someone on a satellite link it's 500ms due to four trips: up to the satellite; down to the server; up to the satellite; back down to the client. And that's the absolute minimum.
Greg Wyler CEO / Founder OneWeb talks about the production of satellites constellation: https://www.youtube.com/watch?...
He talks about speed and latency starting after ~40 seconds into the clip.
Just how low is low orbit? For geostationary I think you get around 250ms latency just to hit the satellite.
Geosynchronous would be the easy way to do it, but a low-orbit constellation is the price of reducing latency to tolerable values. A part of that price will be designing a compact receiver that can deal with two-way communications with moving satellites.
Please, don't proliferate satellites.
The satellites that go up will go down.
Then the satellites can kill you as if it is a meteorite.
Of all the fears about space junk, this one is the silliest. A satellite contains a high percentage of empty space, like a ship. When one deorbits, it burns up. Although s meteor is a solid chunk of rock or iron in contrast, very few of them survive to become meteorites.
With a GEOSTATIONARY satellite. Phased polar and other low orbit assets can be much, much lower. So not the minimum for satellite ping time at all. The minimum is something on the order of 1000km at the speed of light, adding in some distance for practical geometry (the sat one is currently using won't be directly overhead often), I'd guess what, 5ms? The contribution of the protocol and repeater would probably be higher. Remember, light actually travels slower in fiber.
Oddly enough, the FCC is the folks who approve it from an orbital debris standpoint - the thinking is that everyone in space has to have a radio license, so it's a convenient "gate" to enforce the "on orbit life" limit.
These things are up at 700km, so they'll be up a good long time, short of explicit deorbit.
I'm just curious as to whether anyone knows whether 750 miles is low enough to experience enough atmospheric drag to cause the junk from the inevitable collisions to come down in a reasonable time? I personally think it irresponsible to launch satellite swarms of this magnitude at a level above one guaranteed to come down within a couple of years after active station-keeping hardware fails. It seems like I recollect that 750 miles might be above that level.
Are you a civics expert from the space internet?
They're not regulating the space that the signals pass through. They're only regulating the signals.
You're confusing the FAA with the FCC.
Seven puppies were harmed during the making of this post.
No, I didn't even address that mistake. I only addressed the claim that regulation of radio broadcasting isn't the same as regulating the space that the broadcast travels through.
Like the comic https://www.xkcd.com/713/ said, ISS can now get geoip'd, yieldeing ads to "meet local girls in LOW EARTH ORBIT." ;)
Progress is sad