[Starlink] Quick questions about Starlink user terminal

[Starlink] Quick questions about Starlink user terminal

[Sirapop Theeranantachai]

[-- Attachment #1: Type: text/plain, Size: 520 bytes --]

Hi, I'm currently working on packet simulation analysis over large LEO
constellations such as Starlink, and I do have a few questions on the
Starlink commodity terminals and ISL.

1) Is the commodity hardware capable of forming multiple GSL channels to
multiple satellites? If it's not possible, is it possible that the hardware
allows multiple downlink but only one uplink?

2) Does Starlink currently use laser ISL or is it planning in the future?
If so, how is the link performance due to laser accuracy?

Best,
Mark

[-- Attachment #2: Type: text/html, Size: 632 bytes --]

Re: [Starlink] Quick questions about Starlink user terminal

[Ulrich Speidel]

On 12/05/2023 9:48 am, Sirapop Theeranantachai via Starlink wrote:
> Hi, I'm currently working on packet simulation analysis over large LEO 
> constellations such as Starlink, and I do have a few questions on the 
> Starlink commodity terminals and ISL.
>
> 1) Is the commodity hardware capable of forming multiple GSL channels 
> to multiple satellites? If it's not possible, is it possible that the 
> hardware allows multiple downlink but only one uplink?
Based on our observations, no to both questions - one sat at a time. 
We've noticed that power use goes up with observed downlink rate, 
though. Mike Puchol might be able to comment a bit more here as he's 
tried to get between Dishy and the birds.
>
> 2) Does Starlink currently use laser ISL or is it planning in the 
> future? If so, how is the link performance due to laser accuracy?
Laser links are in use, and most likely also between different orbital 
planes, but use is a bit weird. Traffic from Tarawa (Kiribati) emerges 
onto the Internet in Auckland, NZ, 4200 km away and not near a common 
orbital plane in Starlink's dominant 53.2 deg constellation. Link 
performance - how long is a piece of string? We've seen significant 
transient outages from Tarawa, but when there is no outage, rates are 
comparable to what you'd get elsewhere (anecdotally - we haven't seen 
any actual test results from there yet, the owner of the only unit there 
has reported regulatory issues).
>
> Best,
> Mark
>
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink
>
-- 
****************************************************************
Dr. Ulrich Speidel

School of Computer Science

Room 303S.594 (City Campus)

The University of Auckland
u.speidel@auckland.ac.nz
http://www.cs.auckland.ac.nz/~ulrich/
****************************************************************

Re: [Starlink] Quick questions about Starlink user terminal

[Oleg Kutkov]

Hello,

On 5/12/23 00:48, Sirapop Theeranantachai via Starlink wrote:
> 1) Is the commodity hardware capable of forming multiple GSL channels 
> to multiple satellites? If it's not possible, is it possible that the 
> hardware allows multiple downlink but only one uplink?

Regular non-HP terminal works only with one satellite. HP supports 
multiple sats due to a larger antenna field.
The terminal can only do TX or RX with the selected satellite in 
half-duplex mode.


-- 
Best regards,
Oleg Kutkov

Re: [Starlink] Quick questions about Starlink user terminal

[Sirapop Theeranantachai]

[-- Attachment #1: Type: text/plain, Size: 1106 bytes --]

Thank you for very quick responses.
>
> The terminal can only do TX or RX with the selected satellite in
> half-duplex mode.

Just a confirmation, HP terminal can work with two satellites at the same
time, one for only TX, and the other for only RX. Is this statement still
correct?

On Thu, May 11, 2023 at 3:05 PM Oleg Kutkov via Starlink <
starlink@lists.bufferbloat.net> wrote:

> Hello,
>
> On 5/12/23 00:48, Sirapop Theeranantachai via Starlink wrote:
> > 1) Is the commodity hardware capable of forming multiple GSL channels
> > to multiple satellites? If it's not possible, is it possible that the
> > hardware allows multiple downlink but only one uplink?
>
> Regular non-HP terminal works only with one satellite. HP supports
> multiple sats due to a larger antenna field.
> The terminal can only do TX or RX with the selected satellite in
> half-duplex mode.
>
>
> --
> Best regards,
> Oleg Kutkov
>
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink
>

[-- Attachment #2: Type: text/html, Size: 1747 bytes --]

Re: [Starlink] Quick questions about Starlink user terminal

[Oleg Kutkov]

On 5/12/23 01:12, Sirapop Theeranantachai wrote:

> Just a confirmation, HP terminal can work with two satellites at the 
> same time, one for only TX, and the other for only RX. Is this 
> statement still correct?


Correct. It's a hardware limitation. A switch inside connects antenna 
elements to transmitter or receiver chains.
Plus, each Dishy can do TX only in specified time slots. TX slots depend 
on the allowed duty cycle and are defined by a global scheduler.

-- 
Best regards,
Oleg Kutkov

Re: [Starlink] Quick questions about Starlink user terminal

[Sirapop Theeranantachai]

[-- Attachment #1: Type: text/plain, Size: 788 bytes --]

Thank you for the clarification. May I ask how fast one time slot can be
estimately? Is it less than a few milliseconds per unit or is it much
higher? Thanks!

On Thu, May 11, 2023 at 3:18 PM Oleg Kutkov <contact@olegkutkov.me> wrote:

> On 5/12/23 01:12, Sirapop Theeranantachai wrote:
>
> > Just a confirmation, HP terminal can work with two satellites at the
> > same time, one for only TX, and the other for only RX. Is this
> > statement still correct?
>
>
> Correct. It's a hardware limitation. A switch inside connects antenna
> elements to transmitter or receiver chains.
> Plus, each Dishy can do TX only in specified time slots. TX slots depend
> on the allowed duty cycle and are defined by a global scheduler.
>
> --
> Best regards,
> Oleg Kutkov
>
>

[-- Attachment #2: Type: text/html, Size: 1120 bytes --]

Re: [Starlink] Quick questions about Starlink user terminal

[David Lang]

On Fri, 12 May 2023, Ulrich Speidel via Starlink wrote:

> On 12/05/2023 9:48 am, Sirapop Theeranantachai via Starlink wrote:
>> Hi, I'm currently working on packet simulation analysis over large LEO 
>> constellations such as Starlink, and I do have a few questions on the 
>> Starlink commodity terminals and ISL.
>>
>> 1) Is the commodity hardware capable of forming multiple GSL channels 
>> to multiple satellites? If it's not possible, is it possible that the 
>> hardware allows multiple downlink but only one uplink?

> Based on our observations, no to both questions - one sat at a time. 
> We've noticed that power use goes up with observed downlink rate, 
> though. Mike Puchol might be able to comment a bit more here as he's 
> tried to get between Dishy and the birds.

We don't know if this is a hardware limitation or a software limitation, what we 
can observe is that currently it doesn't appear to do multiple connections.

David Lang

Re: [Starlink] Quick questions about Starlink user terminal

[Ulrich Speidel]

[-- Attachment #1: Type: text/plain, Size: 2140 bytes --]

On a side note (TX vs RX):

When people look at the "speed" of Internet connections, there's often a 
focus on "down" as opposed to "up".

For folk working from home using cloud storage for audio or video 
content that they edit, not having a lot of "up" is an issue.

Here's what to expect around our neck of the woods:

https://comcom.govt.nz/__data/assets/pdf_file/0023/314816/MBNZ-Summer-Report-2-May-2023.pdf 
(page 7 and 9)

Now one of the interesting bits about Starlink is that its uplink 
performance is really not that great in comparison to downlink here in 
NZ - pretty much the worst of the current technologies:

This agrees with what I've seen here personally. ADSL in non-fibre areas 
generally now means "miles from the DSLAM", as does VDSL. DSLAM-side 
crosstalk tends to be the limiting factor there.

Non-fibre areas also generally mean "far from the base station". HFC is 
available in only a few select areas here as NZ does not have large 
cable networks.

On 12/05/2023 10:18 am, Oleg Kutkov via Starlink wrote:
> On 5/12/23 01:12, Sirapop Theeranantachai wrote:
>
> > Just a confirmation, HP terminal can work with two satellites at the
> > same time, one for only TX, and the other for only RX. Is this
> > statement still correct?
>
>
> Correct. It's a hardware limitation. A switch inside connects antenna
> elements to transmitter or receiver chains.
> Plus, each Dishy can do TX only in specified time slots. TX slots depend
> on the allowed duty cycle and are defined by a global scheduler.
>
> -- 
> Best regards,
> Oleg Kutkov
>
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink 
> <https://lists.bufferbloat.net/listinfo/starlink>

-- 
****************************************************************
Dr. Ulrich Speidel

School of Computer Science

Room 303S.594 (City Campus)

The University of Auckland
u.speidel@auckland.ac.nz  
http://www.cs.auckland.ac.nz/~ulrich/
****************************************************************



[-- Attachment #2.1: Type: text/html, Size: 3418 bytes --]

[-- Attachment #2.2: e0OmsNLmR6O2npRy.png --]
[-- Type: image/png, Size: 39248 bytes --]

Re: [Starlink] Quick questions about Starlink user terminal

[Eugene Y Chang]

[-- Attachment #1.1: Type: text/plain, Size: 1393 bytes --]

Usually it is deliberate that equipment does not transmit while receiving. This is to prevent the transmit power from overloading the receiver. This problem can be avoided if the TX and RX are on widely separated freaquencies and antenna configurations to keep the TX signal out of the RX signal path.

What do we know about Starlink’s use of laser links for satellite-to-satellite communications? I am most curious about the challenges of pointing the laser at a moving target.
Gene
----------------------------------------------
Eugene Chang
eugene.chang@ieee.org
o 781-799-0233




> On May 11, 2023, at 12:18 PM, Oleg Kutkov via Starlink <starlink@lists.bufferbloat.net> wrote:
> 
> On 5/12/23 01:12, Sirapop Theeranantachai wrote:
> 
>> Just a confirmation, HP terminal can work with two satellites at the same time, one for only TX, and the other for only RX. Is this statement still correct?
> 
> 
> Correct. It's a hardware limitation. A switch inside connects antenna elements to transmitter or receiver chains.
> Plus, each Dishy can do TX only in specified time slots. TX slots depend on the allowed duty cycle and are defined by a global scheduler.
> 
> --
> Best regards,
> Oleg Kutkov
> 
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink


[-- Attachment #1.2: Type: text/html, Size: 8185 bytes --]

[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [Starlink] Quick questions about Starlink user terminal

[David Lang]

[-- Attachment #1: Type: text/plain, Size: 449 bytes --]

On Thu, 11 May 2023, Eugene Y Chang via Starlink wrote:

> What do we know about Starlink’s use of laser links for satellite-to-satellite communications? I am most curious about the challenges of pointing the laser at a moving target.

We know they do it, we see starlink coverage in areas that could not be served 
without such communication. But we have no details of how they do it or the 
problems they solved in creating the system.

David Lang

[-- Attachment #2: Type: text/plain, Size: 149 bytes --]

_______________________________________________
Starlink mailing list
Starlink@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/starlink

Re: [Starlink] Quick questions about Starlink user terminal

[Eugene Y Chang]

[-- Attachment #1.1: Type: text/plain, Size: 1243 bytes --]

The comment (speculation?) about laser communications between satellites in different orbital planes suggest challenges in tracking. I am really curious about the mechanics of tracking and laser pointing.

Gene
----------------------------------------------
Eugene Chang
IEEE Communications Society & Signal Processing Society,
    Hawaii Chapter Chair
IEEE Hawaii Section, Industry Engagement Coordinator
IEEE Life Member Affinity Group Hawaii Chair
IEEE Senior Life Member
eugene.chang@ieee.org
m 781-799-0233 (in Honolulu)



> On May 11, 2023, at 6:28 PM, David Lang <david@lang.hm> wrote:
> 
> On Thu, 11 May 2023, Eugene Y Chang via Starlink wrote:
> 
>> What do we know about Starlink’s use of laser links for satellite-to-satellite communications? I am most curious about the challenges of pointing the laser at a moving target.
> 
> We know they do it, we see starlink coverage in areas that could not be served without such communication. But we have no details of how they do it or the problems they solved in creating the system.
> 
> David Lang_______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink


[-- Attachment #1.2: Type: text/html, Size: 3882 bytes --]

[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [Starlink] Quick questions about Starlink user terminal

[Sirapop Theeranantachai]

[-- Attachment #1: Type: text/plain, Size: 2363 bytes --]

Hi, I have a few follow-up questions

We're simulating the Starlink on a + grid topology on the same inclination
angle.

4 ISLs are static (but delay chages). Two of them connect to other sats
within the same orbital plane, and the other connect to the adjacent plane
(+1 or - 1) .

We know that if the dish is connected to one satellite at the a time, the
latency may not be as great when the orbital planes that the two nodes are
connected to are far apart. It can double the RTT if the orbital planes are
on thr opposite side to each other.


If we have two GSL channels, one of them can connect to the satellite that
moves  north-bound, and the other that moves southbound. This assumption
can provide a significant boost on latency.

Oleg mentioned that HP hardware can do that?
Questions are:
Do you think is this the reason why HP has that kind of antenna?
How did you get that information about HP just in case I need a reference.

Thank you so much!
Mark

On Thu, May 11, 2023, 9:39 PM Eugene Y Chang <eugene.chang@ieee.org> wrote:

> The comment (speculation?) about laser communications between satellites
> in different orbital planes suggest challenges in tracking. I am really
> curious about the mechanics of tracking and laser pointing.
>
> Gene
> ----------------------------------------------
> Eugene Chang
> IEEE Communications Society & Signal Processing Society,
>     Hawaii Chapter Chair
> IEEE Hawaii Section, Industry Engagement Coordinator
> IEEE Life Member Affinity Group Hawaii Chair
> IEEE Senior Life Member
> eugene.chang@ieee.org
> m 781-799-0233 (in Honolulu)
>
>
>
> On May 11, 2023, at 6:28 PM, David Lang <david@lang.hm> wrote:
>
> On Thu, 11 May 2023, Eugene Y Chang via Starlink wrote:
>
> What do we know about Starlink’s use of laser links for
> satellite-to-satellite communications? I am most curious about the
> challenges of pointing the laser at a moving target.
>
>
> We know they do it, we see starlink coverage in areas that could not be
> served without such communication. But we have no details of how they do it
> or the problems they solved in creating the system.
>
> David Lang_______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink
>
>
>

[-- Attachment #2: Type: text/html, Size: 4777 bytes --]