SpaceX Internet Satellite Network: Starlink

[mongo]

If nothing else, sounds like Iridium will continue to have the Antarctic market sewn up

Great circle routes for transcontinental flights will appreciate polar coverage. I think higher altitude Starlink sats will be able to hit all of the US due to FCC coverage requirements so maybe McMurdo will get YouTube...

 

[Cosmacelf]

People here are assuming Starlink will be used for mobile. AFAIK, SpaceX has only talked about fixed terminals. A moving car requires different engineering constraints than a fixed terminal. My assumption is that Starlink won’t be used for marine and all other mobile applications and would indeed be used for fixed terrestrial only.

 

[mongo]

People here are assuming Starlink will be used for mobile. AFAIK, SpaceX has only talked about fixed terminals. A moving car requires different engineering constraints than a fixed terminal. My assumption is that Starlink won’t be used for marine and all other mobile applications and would indeed be used for fixed terrestrial only.

Once you require a phased array to track a VLEO satellite, it's fairly simple to allow that tracking from a relatively slowly moving vehicle. Cruise ships and oil rigs would benefit from bandwidth. Long haul trans continental flights would also, though that requires a faster tracking control loop.

Otherwise, the oceans represent a lot of underutilized capacity.

 

[Bobfitz1]

In addition to all of @bxr140 's fine points, Iridium covers the entire earth pole to pole. Starlink is concentrated in the 55 degree and lower latitudes.

Thanks to you both for taking time to answer my inquiry. One followup. Since Starlink will have a least several competitors, do you and @bxr140 think that starting in perhaps 5 years, satellite internet services are likely to make internet bandwidth much cheaper than it is today? Enough to force the large non satellite providers to follow suit?

 

[mongo]

Thanks to you both for taking time to answer my inquiry. One followup. Since Starlink will have a least several competitors, do you and @bxr140 think that starting in perhaps 5 years, satellite internet services are likely to make internet bandwidth much cheaper than it is today? Enough to force the large non satellite providers to follow suit?

Disclaimer, totally my opinion:
Starlink is going to be great for undeserved areas like me with only DSL, (but at least I have that). It may hit the rural telecoms and ad-hock cell based providers hard. However, it won't impact the general public in major metro areas due to the shear amount of bandwidth needed in those areas.
I think Starlink will be fairly unique in terms of coverage and bandwidth available, I don't see another company being able to compete in the 5-10 year time frame due to lack of rockets to launch a similar constellation and an already tapped market.

 

[bxr140]

Thanks to you both for taking time to answer my inquiry. One followup. Since Starlink will have a least several competitors, do you and @bxr140 think that starting in perhaps 5 years, satellite internet services are likely to make internet bandwidth much cheaper than it is today? Enough to force the large non satellite providers to follow suit?

+1 on @mongo

Short answer is no.

We've talked about it before, but the real problem with Big Satellite Bandwidth--which is the reason the 3 constellations are struggling with funding--is that it ultimately has to compete with terrestrial solutions. Compared to population sparse areas (like The Boons, let alone the third world) which are essentially small islands of individual personal data demand, population dense areas (like city centers) create both personal and commercial demand for data. In those dense areas, which is where the majority of the world's population live, every aspect of a terrestrial solution is cheaper and higher performing compared to a space constellation. IIRC, Starlink is targeting something along the lines of 5G to the end user...many years from now. The corner coffeeshop gives you that and more today for free, to say nothing of what you have access to at your place of work or home.

So...these constellations will make very little impact on how the majority of us (especially in the first world) procure bandwidth, and so these constellations won't make a huge impact to current day pricing. What they will do is significantly enhance the way the rest of the world (the other three billion or...ahem...O3B) access data, and that is absolutely a good thing. The proverbial cure for cancer is brewing in the brain of that 4 year old girl who lives on a remote Indonesian island. Accessible connectivity could even be a catalyst for a migration away from population centers (there's a bit of sociopolitical handwaving on that one, but you get the point).

Anyway, IMHO, and I've said this before, Starlink really comes down to Mars. Not only does it help pay for Mars, it presents a near-complete solution to communication once you're there...

 

[bxr140]

On the competitor front, I missed this last week--the first OneWeb satellites are at launch base now. They're currently scheduled to go on a Soyuz on Feb 19.
FIRST UP Satcom | First OneWeb satellites shipped for launch • Laser ground stations linked to Facebook - SpaceNews.com

This link is in that article, but here's a direct link to the airbus site with two satellite photos.
OneWeb Satellites has shipped first satellites for the OneWeb constellation to launch site
Random observations
--The main antenna arrays are under the crappy looking silver/foil-ish covers on the top of the satellites.
--The solar arrays (the flat panels on either side, covered in grey translucent plastic) are traditional cell-on-honeycomb-panel, which is generally cheaper and easier than the unfurlable solution used on Starlink, generally at the expense of more difficult packaging and more mass.
--Probably the biggest difference between these and Starlinks are that OneWeb has always been lasercomm--that's why Musk and Wyler basically fell out in the first place. So, OneWeb has to use a bunch of gateways on the ground instead of augmenting with ISLs. The two dishes you see on the side of the satellites are for the the gateway antennas.

Also, really interesting comparison between the 3 constellations:
http://www.mit.edu/~portillo/files/Comparison-LEO-IAC-2018-slides.pdf

 

[bxr140]

Edit cause I don’t proofread: starlink is lasercomm, OneWeb is NOT.

 

[mspohr]

The first million are the hardest
SpaceX seeks FCC approval for up to 1M Starlink satellite earth stations

SpaceX seeks FCC approval for up to 1M Starlink satellite earth stations

 

[mongo]

The first million are the hardest
SpaceX seeks FCC approval for up to 1M Starlink satellite earth stations

SpaceX seeks FCC approval for up to 1M Starlink satellite earth stations

Thanks for the article!

One note: article states

A million earth stations may sound like a lot, but it pales in comparison with Dish’s subscriber base of 10.3 million customers for its satellite-based Dish TV service.

However, Dish doesn't uplink to the satellites and are not transmitters.

 

[ecarfan]

From the FCC application, quote: “SpaceX intends to begin launching satellites to populate its constellation in 2019...
In this application, a sister company, SpaceX Services, Inc. (“SpaceX Services”) seeks a blanket license authorizing operation of up to 1,000,000 earth stations that end-user customers will utilize to communicate with SpaceX’s NGSO constellation. These user terminals employ advanced phased-array beam-forming and digital processing technologies to make highly efficient use of Ku- band spectrum resources by supporting highly directive, steered antenna beams that track the system’s low-Earth orbit satellites.”
—————————————————————————————————-

So these “earth stations” are what the customer (the “end user”) will have on their roof, and they will create “steered antenna beams”. So will the stations actually physically move to track the satellite or will the “steering” be done digitally?

 

[mspohr]

From the FCC application, quote: “SpaceX intends to begin launching satellites to populate its constellation in 2019...
In this application, a sister company, SpaceX Services, Inc. (“SpaceX Services”) seeks a blanket license authorizing operation of up to 1,000,000 earth stations that end-user customers will utilize to communicate with SpaceX’s NGSO constellation. These user terminals employ advanced phased-array beam-forming and digital processing technologies to make highly efficient use of Ku- band spectrum resources by supporting highly directive, steered antenna beams that track the system’s low-Earth orbit satellites.”
—————————————————————————————————-

So these “earth stations” are what the customer (the “end user”) will have on their roof, and they will create “steered antenna beams”. So will the stations actually physically move to track the satellite or will the “steering” be done digitally?

I think the phase array does electronic steering

 

[GWord]

I think the phase array does electronic steering

That's exactly right. The concept of a phased array is that you can have some limited beam steering with a stationary array.

 

[BioSehnsucht]

That's exactly right. The concept of a phased array is that you can have some limited beam steering with a stationary array.

Limited may be up for interpretation. Phased arrays can be as few as a few degrees of steering angle or more than 100 degrees depending on the design... Clearly, they're not going to effectively cover a 360 degree sphere, but you the terminals should have no problem with tracking satellites in the intended operation range (I think it was something like no lower than 35 degrees from the horizon, which would be over 100 degrees total steering range), even allowing for non-level user terminals (so needing some extra steering range).

 

[GWord]

Limited may be up for interpretation. Phased arrays can be as few as a few degrees of steering angle or more than 100 degrees depending on the design... Clearly, they're not going to effectively cover a 360 degree sphere, but you the terminals should have no problem with tracking satellites in the intended operation range (I think it was something like no lower than 35 degrees from the horizon, which would be over 100 degrees total steering range), even allowing for non-level user terminals (so needing some extra steering range).

Yes, great point. I was thinking in terms of how arrays like the PAVE/PAWS system or the AN/SPY systems use multiple phased arrays for full coverage where one mechanically steerable antenna might have been used in the past (albeit with lesser capability).

 

[dm28997]

"Taiwan particularly excels in the production of satellites and other related products like solar cells, the news outlet writes. The fact that Taiwan’s satellites cost, on average, one-third to one-tenth of the price of foreign-produced equivalents gives the country a significant advantage in the field." quoted from the link:

Elon Musk's Starlink project 'a good ... | Taiwan News

I wonder how fast investments will come in. Is it possible Starlink becomes public? My thought is after the SEC issues they may want to keep it private. A private placement or perhaps listing on another stock exchange could be feasible.

 

[dm28997]

SpaceX fires up hiring for team to build Starlink orbital broadband network

Hiring for Starlink is ongoing.

 

[ecarfan]

And the Starlink-1 launch is probably just two months away. Will be interesting to see how many sats they pack into the fairing.

SpaceX F9 - Starlink-1 - SLC-40

 

[Cosmacelf]

Bold claim: the fully destructible satellite:

SpaceX Claims To Have Redesigned Its Starlink Satellites To Eliminate Casualty Risks

 

[bxr140]

Boom.
Amazon to offer broadband access from orbit with 3,236-satellite ‘Project Kuiper’ constellation

Apple's next.