C🅰️tSE

Live view of @FCC regulatory staff watching recent Starlink and T-mobile filings asking to be exempt from new rules they helped greate and want to see imposed on others.

🧶🐈‍⬛

1/n

In 2020 T-mobile asked FCC not to allow ASTs system w/o filings to show that they do not cause harmful interference to terrestrial networks.

$TMUS also requested new rules that FCC started.

AST showed their system doesn’t cause interference. ASTS design criteria since 2018.
2/n

In 2022 T-Mobile and Space-X Starlink
recognixe that direct to cell will be a reality.

Essential for T-Mobile to stay  competitive &amp; much more profitable than Starlinks legacy VSAT architecture / fixed internet protocol service.
They join forces and declare d2c cooperation.

3/n

- CatSE hold on, aren’t you being a bit harsh on Starlink d2c v2 minis?

Why?

- That’s a band-aid, Cat. 🩹

Right, so maybe a bit harsh. 

But, they do look like that. HEO Robotics published an image of one of the SpaceX Starlink V2 Minis deployed.

- So where’s the array?

4/n

To use the same analogy of band aids. Starlinks have a more conventional design.

It restricts the area of the phased array.

This smaller phased array area imposes severe technical limitations on what the v2 minis are capable of.

-Fewer antenna elements fit?

Yes.

5/n

- T-Mobile first protested the novel formfactor startup asking the FCC to make harsh rules to contain ASTs disruptive effect and later joined forces with the less technical advanced contender?

-Yes.

And now those same rules comes back to bite them?

-Yes AT&amp;T and VZN files.

6/

- Can we get back on the technical limitations?

Right. A smaller array allows for less directive beams creating wider cells and a wider area around the beam is affected by interference from that main beam.

-Lots to digest there.

That’s just the starter. I’ll elaborate.

7/

To understand the main beam and importance of directivity we can use some radiation humans can relate to.

Visible light.

Here is a flashlight beam and a laser beam. The latter is more directive.

And so is the splash.

-Right. Now I get the starter. What’s for main dish?

8/

Let’s just dwell on the two implications of the higher directivity on Area Spectral Efficiency.

IF both systems have the same bandwidth available for a cell. Then the users of that cell gets to share the same bandwidth.

This means higher throughput per user in small cells.

9/

-Right. And the second effect?

You create more narrow cells by higher directivity. Like laser light is stronger than a flashlight not by bigger battery or more energy, but by means of more focus.

-High signal strength?

Yes. And that allows for higher order modulation.

10/

-So, let me sum up the starter:

- Higher directivity means:

  1) smaller zone affected by main beam splash / integerenge while you also
  2) get more bandwidth per user at 
  3) higher signal strength and higher order of modulation. 

Yes.

-Wow.

11/

-Your previous slide said ”Signal to Noise” not signal strength.

Yes. It’s a bit like when you try to get heard in a noisy room. Other peoples signal becomes noise to you.

And you need to raise your voice to punch through.

- like at a party?

If so Starlink are noisy kids.
12/

- They want to raise the noise floor?

Correct. And this brings us over to the main thing. 

Aggregate interference caused by sidelobes and main beams from a massive constellation like Starlink d2c on other satellite systems and on MNOs

-Sidelobes?

Yes. Big difference.

13/

The higher directivity also reduces sidelobes in relative terms.

-So this is not the splash from main beams?

No it’s ”splash” from unwanted ”side-beams”.

And AST lowers that in two ways. By higher directivity and that patent on previous link.

/14

Starlink doesn’t have these tweaks and they rather see the new rules for the allowed PFD  level of their noise bent by means of waiver or broken by means of a petition to change the rules.

AT&amp;T and Verizon sees how such action would degrade their terrestrial networks.

/15

AT&amp;T has PCS block. The block Starlink d2c has emissions in.

Makes them an affected party.

-So, T-Mobile isn’t just hurting their own subscribers?

No. 

AT&amp;T will fight them.

/16

-Can’t FCC allow the 😎 cool Space-X company to spam terrestrial networkd more than AST, after all they make 🚀 macho rockets?

AST has ocercome a lot of risks, among them regulatory risks whereby they received a license.

Key to that: DO NO HARM.

- Did you mean to say ”overcome” and 
how do you know that Cat?

Yes. Try typing on a phone with paws.

Regulators ultimately strive to maximize the common good and minimize harm.

Existing incumbents delivering valuable services enjoy massive protection by regulators.

/18

And the way regulatory staff operates is well described here.

There is this principle.

- Which one?

It’s called ”BAT”

-How do you know this?

I’ve chaired regulatory boards for more than a decade.

-You. A cat?

Sweden is an advanced society and I’m a Cat, with a BAT.

/19

-Tell me aboout BAT.

While not directly picking fight with Starlink, (AT&amp;T, Verizon and Omnispace will do that just fine), AST did a filing which T-Mobile asked for vocally.

One re: interference.

In that they showed AST tech can use PCS block without harmful interference.
/20

Tge way regulators work in a situation where they need to balance doing some good against doing some harm is to apply the BAT principle.

- They require an applicant to use the best available technology to minimize harmful interference?

Yes.

-And what is that BAT benchmark?
/21

There are 5-6 interference filings from AST of interest.

In particular one in 2020 showing compliance with protecting terrestrial operations and one in 2024 showing that AST doesn’t cause harmful interference to satellites either.

-So AST has shown their tech is BAT?

Yes.

/22

- I still don’t quite see that a few Starlinks could do so much damage.

Let’s cycle back to this.

See how SIR is proportionally better with AST?

To compete Starlink wants to up the amp.

-Aha! Whereby they also increase the destructive interference from sidelobes!

Yes.

/23

- So. Signal to interference ratio. SIR is key?

Yes.

Spectrum is defining  scarce resource of our age. The information age. BAT applies.

And then there is this thing.

-Planet earth?

Yes. Few of those. Also a scarce resource. We need to be good stewards of it.

-Right.

/24

Here is planet earth with a few thousand Starlinks in Low Earth orbits. Schematically.

-Your point being they would be many?

Yes. You said a few. I think ”quite a bunch” would be more appropriate. Starlink wants a mega constellation.

- Interference aggregates?

Yes.

/25

- I’m full. Dessert?

ESG replenishing Starlinks constellation as it has worse orbital dwell time,

Elon endorsing Trump while  FCC has democratic majority,

Cyber security &amp; Lawful interception in regenerative architecture,

6G OTFS &amp; stealth coms ?

Or

Likely outcomes ?

/26

- Likely outcomes, with coffee. Thx.

The likely outcome is that FCC denies Space-X request to lower the protection. Whereas they might have reduced that some after trial data. They will not allow such massive interference.

It will lower Starlink d2c throughput.

-How much?

/27

By a lot. More than cut in half. Starlink will be good for text services.

But they are fast learners.

This means effort will shift to other bands and bigger satellites. They will seek to catch up to ASTS tech.

- And T-Mobile?

They will likely use Starlink for texts.

/28

- Starship is that needed for those bigger Starlink satellites?

It will help. V2s can improve some within a Falcon fairing, but not much.

-And AST?

They are minding their own relentless innovation called Block2 satellites and what is beyond that.

- What is beyond?

/29

AST placed their G agnostic Nokia AirScale single RAN gNodeBs terrestrially so that delay and doppler compensation can also be terrestrial.

It allows for multi sat MIMO and OTFS multi signal paths.

6G tech.

-So Starlink will try to catch moving target?

Yes. With a twist

/30

- I saw what you did there, Cat.

Band aids to symbolize Starlink V2?

-Yes. Not kind. And what is the twist?

Picture a race.

One contender starts early, in 2018, from first principles.

Plans everything chooses the path wisely. And starts down the choosen track.

-Gotcha.

/31

Starlink d2c went low and many because their second grade tech does not allow to go high and few.

It is a mind blowing exercise to contemplate the ”What IF” of $ASTS going low and many because they opt to do so.

1/n 

Let us cut altitude in 2.

First of all Capital Expenditure would increase by 4x from close to nothing towards a tiny fraction of global tower CapEx.

Second. Area Spectral Efficiency would   More than 15x due to:

1. 4x Better SNIR/better link budget
2. 4x More spectrum/ user
2/ 

Third. Most rural towers would become obsolete.

Fourth.  $ASTS would become top MC company of this particular planet and The network to connect most people and machines.

3/ 

”What IFs” are interesting thought experiments.

I’m bullish as can be on $ASTS even w/o going far down such rabbit holes.

You could probably have seen bearish what-IFs  and there is this flip side to them:

Things can also go _better_ than publicly announced.

4/n 

This thread was brought to you by a swedish cat. 🐈‍⬛ 🇸🇪

Spoil your local animals; Pat a pet.

This is why there are weekends.

And as always, I promise to bite a short for every retweet.

😺🫎

5/5 

🚨SPACEMOBILE $ASTS FULL FCC GRANT FOR 🇺🇸 BLOCK 1  β-TESTS🚨

✅$ASTS REGULATORY READINESS NOW ON PAR WITH STARLINK D2C 

✅$ASTS TECHNOLOGICAL READINEDS WAY BETTER,

⏳ WHICH IS ABOUT TO BE SHOWCASED 🎬🎥🍿

🧶🐈‍⬛

1/n 

As a ten tweet primer the competition in the form of Starlink regenerative architecture(green👇) is having severe problems with handovers.

As also can be seen AST is a different architecture altogether.

You’ll notice:

One single core.
gNBs (5G)

V/S

Two cores
eNB &amp; gNB 

2/n 

But it is far from the only differentiator.

Here is a primer to another difference needed as background.

Starlink uses earth moving cells, EMC.

AST SpaceMobile uses earth fixed cells, EFC. Picture hex grids all over the planet each cell having it’s own cell ID.

Fixed.

3/n 

This means AST has much fewer handovers.

And they are satellite to satellite handovers like the cell highlighted here.

AST trains and aims their beams on a fixed target.

Staying on target.

4/n 

Starlink is very different.

Firing ftom different cells as they drive by in rapid succession.

This means your cell ID is constantly shifting as the very rapid VLEO satellites sweep close by without using adaptive beamforming.

Without training their beams on target.

5/ 

We saw this was coming, note:

”..but the Cell-ID# kept changing!…”

And I have pointed out this fundamental design flaw of Starlink d2c for quite some time.

6/

Here is a video of a very recent beta test of the current Starlink.

Experiencing repeated dropped connections ✂️on rapid fire beam handovers. 

https://www.youtube.com/watch?v=vD6gzS7DP94

It’s not looking good.

7/

More differences exists, such as much better signal quality in $ASTS system.

By being regulatory compliant on out of band emissions AST may use full power.

Interestingly this better RF tech also prevents self interference.

But lets cycle back to the architecture schematic.

8/ 

Notice NOKIA here?

Nokia has a substantial patent portfolio, which is a key asset for the company

Nokia has a total of  56,945 patents globally, 26,281 patents of which granted. 20,000 patent families, with over 7,000 declared as essential to 5G.

Nokia Partner is stronk.

9/ 

Their US research laboratory, Nokia Bell Labs, celebrates 100 years of shredding pesimistic bears worldview to pieces by advancing communications technology.

They have been involved with $ASTS refining AST technology for three years now.

It matters.

10/ 

It matters when we research the foundations of ASTs terrestrial component (which Nokia provides) and their ability to do seemless handovers etc.

The unified network architecture, years of collaboration, all that tech built for this. It all helps.

11/ 

Zooming in on $ASTS architecture it is these terrestrial gateways and their feeder link that now has been approved.

It was the last outstanding permit needed.

And the Block1 system already tuned in UK and Turkey is now ready to go live in the US.

12/ 

🥇THIS IS MUCH BETTER TECHNOLOGY THAN WHAT STARLINK IS CURRENTLY FIELDING. 

⚔️AND THERE IS - STARTING TODAY  - NO LONGER A REGULATORY LEADERSHIP FOR STARLINK ON The US MARKET. 

🥇AST LEADING IN BUSINESS AGREEMENT 

13/ 

In the words of @thekookreport it’s a Trump approval.

Brendan Carr @BrendanCarrFCC  seems to be a guy to get things done.

And todays approvals leaves the shorts spreading FUD on him being biased towards Musks network in shame.

Because they were very wrong as we shall see.

14/ 

The grants were of these two filings.

It is noteworthy, as these two were the long term ones.

Put on public notice for a 30 day comment period.

Often a much more lengthy process.

15/ 

And guess what?

Brendan Carrs FCC approves it exactly one month after those filings went on Public Notice.

Not one single day of delay beyond what is required in regulations.

16/ 

Approving these long term grants so quick means the pending shorter period gateway applications for same spots aren’t necessary anymore.

These grants are not the 60 days we expected they are for 180 days extending well into the period of full SpaceMobile grant incl. Block2s

17/ 

What to expect?

✅THESE GATEWAYS ARE ALREADY IN PLACE

✅SPACE SEGMENT ALREADY TUNED IN UK &amp; TURKEY

✅✅ MULTIPLE 10 MHZ CHANNELS MEANS WE EXPECT TO SEE HIGHER THROUGHPUT FROM THESE TESTS.

18/n 

Worst case is 4 bps/Hz as was accomplished  with BlueWalker 3 but on twice the bandwidth 10 MHz resulting in 40 Mbps downlink:

Best case from these test results involve stuff like using dual polarization, dual 10 MHz channels or a combinationof those to even higher throughput
🏁

King Tut outlines the PR impact $ASTS partners can make. 

Read it!

It is an stated objective by AST to PR their achievements through their partners, as could be seen recently when Vodafone CEO did d2c video call in UK.

- More to come!

By advice of @spacanpanman I add an servicelink phased areay size comparison. 

Comparing a cpuple of recent press-releases we see PNT mentioned in the first March 06 2023.

The teaming agreement with Fairwinds.

2/n 

Going further back in time we find an AST pitch submitted 14.04.2020 with AfWerx. (US AirForce).

3/n 

Interesting read.
It mentions PNT in some words.

4/n 

Did AST get the DoD interested in 2020?

We do know that AST executives has an ongoing R&amp;D effort in PNT.
See patents.

There are several benefits from combining LEO + advanced Delay doppler managment + beamforming + communicating correction signals + round trip time measure.
7/n 

Several advanced techniques to do alt PNT is embedded in 5G advanced standards. So there is a civilian standard for this, besides the DoD interest fueled by weak legacy GPS being challenged by many threats.

8/n

The only application ever filed by AST &amp; defense, later AST SpaceMobile is to mount an GPS repeater antenna inside Midland main hangar.

I’m going to show it in parallell with an L3 Harris application for the exact same gadget.

AST left. L3Harris right.

9/n 

We do know L3 Harris makes PNT satellites. (And a lot of coms gear).

But the purpose of both companies to recreate outdoors GPS signals indoors in 2021 might just be to do Telemetry Tracking and Control experiments.

Still, these are similar no?

10/n 

Maybe just great minds thinking alike in 2021?

But the similarities in the wording seems to suggest a connection of some kind between the two companies.

Beyond the fact that the AST filer Midland manager Ken Kramer used to work at L3.

11/n 

The timing of AST installing a GPS repeater in Midland main hangar is summer of 2021.

A year after the AFWerx pitch. 
A year before launching BW3.

In the year that followed there was a shift in strategy.

Announced August 2022 prior to the delayed BW3 launch.

12/n

AST opted to postpone a few large ASIC block 2 test satellites and prioritize an intermediate step of many FPGA (software defined) intermittent service satellites.

Needless to say The FPGAs are more flexible for more use cases incl broadcast. While less capacity for coms.

13/n 

That shift of strategy might have been caused by alt use case doors opening up.

It may also have been influenced by financial concerns. How to reach additional revenue streams from the tech early.

And if so, what alt use cases likes the FPGA design? PNT perhaps part of it?
14/n

In the 2020-2021 period the president issued policy on GPS.

Instructing on january 15 2021 several government entities to take actions to improve PNT.

https://www.gps.gov/policy/docs/2021/

Did AST adapt in 2021-22 to this demand for fused / cooperative LEO PNT as a dual use case?

15/15 