I know that in the 90s, a company called Teledesic was going to make a huge LEO constellation to offer satellite broadband, but now it seems that every satellite broadband offering is geostationary, with the attendant high latency issues. Is there something inherently inadequate to the purpose at lower orbits? Or is it that none of the LEO operators happen to have the right spectrum?
As an aside, come to think of it, is there really a future for satellite internet? If we use rural electrification as a precedent, doesn’t it seem more likely that the government will eventually create the right incentives to get fiber/towers to all the remote, underpopulated areas?
With a geosynch satellite, you only have to point your dish in one direction. Any other orbit, and you’d have to track it across the sky somehow. There are ways to do this with no macroscopic moving parts, but it’s still a lot easier to just have a single dish pointed in the right direction.
The LEO Iridium system needed 66 plus spares satellites. That’s a lot of launches. Plus, yes, if you have a fixed spot in the sky you can use an aimed dish directional antenna and get by with less power (longer battery life) IIRC one of the suitcase phones had to have a self aiming dish?
Because a low orbit satellite is only in view for a few minutes at a time you need dozens of them for coverage, ergo its *phenomenally *expensive! There isn’t a big enough potential market for rural broadband to make it anywhere nearly financially viable.
Iridium itself went bankrupt soon after going online, after over a decade of development and billions invested. In that time cellphone coverage had expanded so immensely that satellite phones were practically obsolete! Motorola was days away from detasking the first group of satellites (moving them out of active orbital paths, to eventually let them re-enter and burn up) when the US military took them over.
That, and good ole’ fashion supply side economics will eventually see pretty near the whole US wired. My brother had DirectPC years ago, when it was his only high speed option. He had the original one-way, dial-up return system. It was high speed internet, but not without some annoying drawbacks. Best thing I can say about it is that it *was *better than dial-up.
To be clear, the launches of Iridium satellites were all multiple deployments (usually six deployments in the payload stack). But at $200M-$300M a launch on a Delta II, it was still a hefty bill for an uncertain market. And while Iridium had a relatively robust scheme for handing off voice calls, I think trying to do that with a connection using any kind of virtual private network (VPN) protocol might be challenging without a very complex and robust network aliasing scheme. But mostly, there just isn’t much demand for such a service.
Pace other commenters, isn’t there demand for such a service? There’s something like 20mm homes without access to broadband in the US alone, and many many millions more around the world (since this would be a network with global coverage, which could potentially lease out transponders to local service providers in other countries).
So I still don’t see why such a system doesn’t exist. Say it takes 66 satellites at a total cost of, I don’t know, $10 billion (probably much less–Iridium’s constellation cost $6 billion). If you get 20 million people paying $30/month for internet, that would be over $7 billion in (very high margin) annual revenue for 10 years (the useful life of the satellites). Moreover, if this would be carried on the ka band, it would be a higher quality connection than you could get from a terrestrial wireless network, possible attracting high-end users (like airlines, ships, oil fields, etc).
The economics seem to make sense. Is there some technical reason that makes it impractical, like the fact that connections would have to continually be handed off from one satellite to another as they move out of range of the receivers? Iridium’s phone system seems to be able to deal with that challenge, but I don’t know if internet access is just a different ballgame from a technical standpoint.
IIRC, the US took over Iridium because of a technicality. The Iridium company went bankrupt - and wanted to write off their investment in the satellites to recover. According to US accounting law, to write off the satellites they had to dispose of them (othewise they still have a value). Nobody would buy them, so they were about to de-orbit them to satisfy the accountants, when the military decided to take them rather than waste the technology.
A fundamental problem with these systems is that they are predicated on providing service to places where there aren’t many customers, but where those customers are willing to pay more money for an inferior service. It hasn’t turned out to be a great business model.
Actually, the reverse. We’re nearing the limits of what companies find profitable in terms of wiring the country for broadband.
Basic supply/demand means that it’s too expensive to run wires to some of the more remote locations in the country to serve only a few customers. It’s one reason why postal service is one example of something the government actually does well. Private enterprise would not find it profitable to offer regular, cheap mail service to East Podunk, Montana.
Rural electrification is a good example of this principle. So is universal landline telephone coverage. There are still a few small towns that don’t have any phone coverage (but are often but not always close enough to a cell tower to get phone connectivity that way). If it’s too expensive for private enterprise to get low-bandwidth phone lines to rural areas, it’s also probably too expensive for them to get broadband to the same folks.
The FCC is looking into transferring the universal service fund (the fund that helps pay to get landline telephony to rural areas) into a fund that pays to get broadband access to rural areas (link).
Even if the demand is there the typical economic and income demographics of those in outlying, non-served areas (ie many are very, very poor) in the US does not bode well for being able to generate a useful cash flow relative to the cost of the infrastructure.
Also, people who *choose *to live in very isolated areas tend to like things that way. Which makes me think they’re not exactly going to be clamoring for better broadband internet. Sure some are, and for those there’s current, geostationary satellite internet. The only advantage that low orbit satellite internet would bring is less latency. Big deal. That’s hardly a reason to invest billions and billions of dollars!
I do know some people who live in isolated rural areas who are desperate for high speed connectivity. They love living where they live, but would also love to be able to watch YouTube, upload video clips, etc. They have looked at Hughes’ satellite system, but installation is very expensive - being a transmission system, installers have to have a pretty technical FCC license.
I have another friend who has an Iridium phone. He’s a sea kayaker, and it is the only tool that will work. The flaw in Iridium’s business plan was believing executives self-image as vastly important people who had to be in constant contact, forgetting that they are always in places with regular cell phone service.
So much bad information here I had to register an account just to counter some of it. This site says it fights ignorance. As far as I can tell from this thread its actually spreading it.
First of all, this is not that expensive to build. LEO satellites are not subject to the radiation of high orbit satellites. So they can be built inexpensively with commercial grade parts.
Secondly it is not that expensive to launch these satellites. They are small and light and in low orbit so you can launch multiple satellites at one time. Possibly anywhere from 12 to 24 in a single launch.
Thirdly, non-geostationary satellites should not be a problem. We already have a system that does this in reverse. A cell network. When you are driving you do not lose your data connection when the phone switches towers. The difference with a satellite network is that it’s the satellites moving instead of the users.
It may have been challenging and expensive when Teledesic proposed it but it’s not anymore.