x
Services

Musk's Starlink is 'not some huge threat to telcos'

WASHINGTON – The way Elon Musk sees it, he'll be helping telcos by taking hard-to-reach customers off their plates with broadband services that will soon be offered by SpaceX's Starlink business.

The Starlink project aims to use low Earth orbit (LEO) satellites as a delivery mechanism for relatively high-speed broadband services to remote areas. SpaceX has secured licensing from authorities to launch more than 12,000 satellites into orbit and the company said it will be able to increase Starlink broadband speeds it provides with each satellite launch. Starlink's target markets include the Northern US and Canada this year, "rapidly expanding to near global coverage of the populated world by 2021," according to the Starlink website.

SpaceX CEO Elon Musk, being magnanimous to telcos.  (Image courtesy of the Satellite Show 2020.)
SpaceX CEO Elon Musk, being magnanimous to telcos.
(Image courtesy of the Satellite Show 2020.)

At the Satellite 2020 conference here on Monday evening, Musk said the global business of providing Internet access to the underserved could be, "as a rough approximation," as high as $30 billion a year. "Starlink is not some huge threat to telcos. I want to be super clear: it is not," Musk told the crowd here.

On the contrary, it could be helpful to telcos by serving populations not well covered by cell towers, copper or fiber, Musk said.

"5G is great for high density situations," Musk said, "but it's actually not great for the countryside, you know, for rural areas. It's not great; you need range. And so in any kind of sparse environment 5G is really not well suited."

"So Starlink will effectively serve the three or four percent hardest to reach customers for telcos, or people who simply have no connectivity right now. Or the connectivity is really bad. So I think it will be actually helpful and take a significant load off the traditional telcos," Musk said.

Musk said Starlink is aiming to provide broadband service with latency "below 20 milliseconds," fast enough for a "fast-response video game at a competitive level."

On the ground, Musk said the user equipment will be a sort of "UFO on a stick" and users will point it at the sky and plug it in. Those instructions can be done in either order, he joked, his point being that Starlink wants to make the experience dead-easy for consumers.

By definition, a company providing connectivity that's easy to get and reaches anywhere has to be a huge concern to telcos and cable providers – companies designed and run for decades to exist as unchallenged regional monopolies.

Musk's Starlink will get competition in space. Several companies are investing in satellite broadband opportunities, including a new joint venture from Vodafone Group and Rakuten, SpaceMobile, that was revealed earlier this month.

While Starlink is looking to provide hardware on the ground that's easy to set up and configure, SpaceMobile and others are looking to deliver service that will allow for "seamless roaming to and from terrestrial cellular networks at comparable data rates without any need for specialized satellite hardware," the group's press release claims.

Starlink, SpaceMobile and others will be targeting different geographies and customer sets, but the overall effect on telcos could be similar: parts of a $30 billion global connectivity market (using Musk's best guess) could disappear from the telcos' addressable market.

Lest anyone think that SpaceX and Starlink will be the regional telco of the skies, Musk said SpaceX, according to its business model, is also helping other companies launch their satellites. He said he doesn't see an issue with helping firms competitive to Starlink establish their space infrastructure so they, too, can compete. "We're giving them a good deal, by the way," Musk said of SpaceX's ability to help satellite constellations get into orbit.

Musk added: "The world seems to have an insatiable appetite for bandwidth. So we're certainly happy to launch other satellites and, you know, we don't think Starlink is going to destroy all the other satellites or something like that – definitely not."

Phil Harvey, US Bureau Chief, Light Reading

ChiefEng45876 3/11/2020 | 7:16:11 AM
Re: Isn't Satellite Internet really the target? I suspect Musk wants to do both -- the business thing of making lots of money out of the financial services industry to finance his Mars plans, and the idealist thing of providing low-cost internet access to places which don't currently have it.

The contention issue is no more of a problem than in any network, you partition the system either by bandwidth or capacity so high-priority traffic always wins -- typically this is done by allowing much higher latency and buffering delays for the low-priority traffic, which doesn't need millisecond latency anyway.

Anyway the antennae in systems like this are phased-arrays which can steer multiple beams to multiple terminals simultaneously (TX and RX) without any contention. Depending on how the signal processing is done they could also partition the beams by frequency, or the available bandwidth could be split between low and high priority traffic as you suggest, but this tends to reduce the average available bandwidth for high-priority traffic.

Traffic prioritisation is really not an issue, high-priority business will pay for the system and get whatever bandwidth it needs, rural internet access will get whatever is left over essentially for free. The existing internet could also do this perfectly well if not for the fact that such prioritisation is legally banned in many places, Musk won't have this problem as the owner of a private network...
brooks7 3/10/2020 | 7:22:44 PM
Re: Isn't Satellite Internet really the target? More than priority.

Think of it this way.  In an environment with many ground stations sharing the same uplink, there is delay in accessing the bandwidth to allow for a contention protocol of some sort.  The way around that is to have separate frequencies for the different services.  The thing I would wonder is why you will put all the complexity of serving a 1 to many environment and combine it with a 1 to 1 environment.  One would think that if the money is all in the 1 to 1, then that is what you would do and skip the rural bandwidth.

seven

 
ChiefEng45876 3/10/2020 | 12:21:05 PM
Re: Isn't Satellite Internet really the target? From memory the transmission overheads add about 10% to the delay (from London-NY), including all delays the time saving was still 3x bigger than the straight-line fiber.

Obviously the high-priority-financial-pays-for-Starlink traffic will have to take priority over low-priority-rural-internet-piggyback traffic, and Starlink will have to make this work -- which I'm sure they will since their whole business depends on it.
brooks7 3/10/2020 | 11:26:24 AM
Re: Isn't Satellite Internet really the target? So the packet collection and retransmission becomes the big effect then, depending upon the bit rate.  The other thing that would be odd is that if I was going to build that then you need only cover some major metro areas.  You would not want to mix that kind of traffic with residential traffic, even if it was prioritized.  You would want to have dedicated frequencies so that there is no multi access layer involved to slow things down.

seven

 
ChiefEng45876 3/10/2020 | 11:17:17 AM
Re: Isn't Satellite Internet really the target? Yes, because light travels in a fiber at about 2/3 of the speed it does in air or vacuum. Even including the encoding/decoding delays the time saved is more than even a straight-line fiber.
brooks7 3/10/2020 | 10:43:08 AM
Re: Isn't Satellite Internet really the target? So all that distance (which is what is the time involved) is better than a direct fiber connection like the low latency connection guys have done?

seven

 
ChiefEng45876 3/10/2020 | 10:30:44 AM
Re: Isn't Satellite Internet really the target? Musk isn't aiming to make loads of money from providing satellite internet to remote areas, the cash cow for Starlink is high-frequency trading because sending data up to a satellite, across to another one through space (actually, can be multiple hops) and back down again is faster than sending it through optical fiber.

This is the market where every millisecond is so important that they already spent half a billion dollars laying a straightline fiber from London to New York instead of the normal wiggly one. Starlink will save (from memory) 3x as much delay as this, and provide this service to every financial centre across the world, with the time saving being bigger the further thay are apart.

Providing low-cost remote satellite Internet access will piggyback on the network paid for by financial trading, it doesn't even have to make Musk any money. IIRC the estimate is that a terminal will cost about $200, which is way cheaper than any other way of getting data out to the sticks.
brooks7 3/10/2020 | 9:39:41 AM
Isn't Satellite Internet really the target?  

I mean the way I put it to people is that every location in the US can get Satellite Internet from one of two vendors today.  Now it is pricey and has limitations.  And there are actually locations in mountainous where it doesn't work.  I think we need to see how this competes in the US and other industrialized countries versus the 3rd world.  I was under the impression that we are talking about 3rd world here.  My only question then is about electrical power and the cost of computing.  Do rural 3rd world folks have regular power and have the money for computing?  They definitely do for phones.  

That has always been my question about these projects.  What is the TAM for real?  Not the number of people that they can cover.

seven
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE