SpaceX COTS Demo 2/3 Updates (THREAD 2)

[wolfpack]

From ISS On-Orbit Status Report for 11/12/2011.

CUCU Testing:
Ground-commanded loopback testing of CUCU (COTS UHF Communications Unit), to be used for the first SpaceX Dragon Demo in February next year, took place via S-band on 12/09 & 12/10 between 11:00 PM GMT and 3:00 AM GMT, performing transmit/receive tests between CUCU 1a and CUCU 1b. The goals of these tests were (a) to prove communications functionality over combinations of CUCU strings and ISS antennas as proved through demonstrating very low (<1e-05) BER (Bit Error Rate) in at least one transmit power setting, (b) to gather data that could potentially distinguish whether behaviours observed in the prior day's tests are indicative of CUCU-internal issues or issues with cabling or connectors downstream of CUCU transmitters, and (c) to characterize CUCU performance as a function of transmit power.

1e-05 BER is atrocious. Maybe a misprint? Most comms links like 1e-9 or better (HDMI 1e-9, Gigabit Ethernet 1e-12, etc).

[Antares]

And those are physical links.  How long is your cable?

[ugordan]

1e-05 BER is atrocious. Maybe a misprint? Most comms links like 1e-9 or better (HDMI 1e-9, Gigabit Ethernet 1e-12, etc).

Cable or wireless?

[wolfpack]

1e-05 BER is atrocious. Maybe a misprint? Most comms links like 1e-9 or better (HDMI 1e-9, Gigabit Ethernet 1e-12, etc).

Cable or wireless?

Wireless is worse, but 1e-5 is one bit error every 100,000 bits. What's the baud rate of the link? If it's near 100k baud, then you're making an error every second. That's fairly useless as a comms channel. If the baud rate is much, much lower then it's OK.

[pippin]

Depends on the conditions. I have worked with much worse comms channels, you need to add application layer error correction but it's not like that hasn't been invented.... You will probably not use it as a high-speed internet link but for a comms link under unknown or varying environmental conditions and moving trasceivers it's not THAT bad.

[ugordan]

Wireless is worse, but 1e-5 is one bit error every 100,000 bits. What's the baud rate of the link? If it's near 100k baud, then you're making an error every second. That's fairly useless as a comms channel. If the baud rate is much, much lower then it's OK.

They're smart people, if they didn't expect BER of 1E-5 to be adequate, they would have designed it differently.

[Comga]

  The location of the ISS in the orbit is not critical,

Yes, it is critical. 

OK "Critical" was the wrong word.

What was meant was that for any given ISS location around the orbit at the moment of launch, a rendezvous can be designed by adjusting the orbit height of the approaching visiting vehicle as a function of time.  Is that true or not?

[simonbp]

They didn't seem to require that on Delta III's maiden flight...

Maybe not the best example! 

[ugordan]

They didn't seem to require that on Delta III's maiden flight...

Maybe not the best example! 

That is precisely why I brought it up. You would think the 3rd flight would be less iffy than a maiden launch of a vehicle. Yet they still launched D-III at night.

[Jim]

  The location of the ISS in the orbit is not critical,

Yes, it is critical. 

OK "Critical" was the wrong word.

What was meant was that for any given ISS location around the orbit at the moment of launch, a rendezvous can be designed by adjusting the orbit height of the approaching visiting vehicle as a function of time.  Is that true or not?

It depends on the chasing vehicle.  For the shuttle, no, it would take too long.

[Comga]

  The location of the ISS in the orbit is not critical,

Yes, it is critical. 

OK "Critical" was the wrong word.

What was meant was that for any given ISS location around the orbit at the moment of launch, a rendezvous can be designed by adjusting the orbit height of the approaching visiting vehicle as a function of time.  Is that true or not?

It depends on the chasing vehicle.  For the shuttle, no, it would take too long.

But for Dragon?

Specifically, is there a signficant probablility that the ISS is in such a poor place around the orbit that launch would be scheduled for the next or subsequent day to get a better plan, even an earlier arrival?
Is this a normal part of launch scheduling?
Did the ISS altitude raising take this into account for the Shuttle?

[Jim]

yes, see phase angle

[deltaV]

1e-05 BER is atrocious. Maybe a misprint? Most comms links like 1e-9 or better (HDMI 1e-9, Gigabit Ethernet 1e-12, etc).

I wonder if 1e-5 is the error rate before or after error correction? See e.g. http://en.wikipedia.org/wiki/Error_detection_and_correction#Deep-space_telecommunications . I suspect that the 1e-12 rate you quoted for ethernet is probably the rate of errors that sneak by the built-in error correction. If 1e-5 is the error rate before error correction then the error rate after error correction would be too close to zero to worry about (with appropriate choice of codes). If 1e-5 is the rate of undetected errors after error correction then I would conclude that SpaceX must be sorely lacking in talented electrical engineers and computer scientists. If 1e-5 is the rate of errors that are detected but not corrected (except by resending) that's probably tolerable in this application.

[Robotbeat]

How can there even be a question if that is the BER before or after correction? You wouldn't design a system which requires high reliability that has that many uncorrected errors. At 10^-5 BER, it is not a terribly great challenge to eliminate basically all errors using an appropriate error correction scheme. So obviously that's the rate BEFORE appropriate error correction.

[deltaV]

How can there even be a question if that is the BER before or after correction? You wouldn't design a system which requires high reliability that has that many uncorrected errors. At 10^-5 BER, it is not a terribly great challenge to eliminate basically all errors using an appropriate error correction scheme. So obviously that's the rate BEFORE appropriate error correction.

I agree that it being before correction is far more likely than after for the reasons you state.

[wolfpack]

1e-05 BER is atrocious. Maybe a misprint? Most comms links like 1e-9 or better (HDMI 1e-9, Gigabit Ethernet 1e-12, etc).

I wonder if 1e-5 is the error rate before or after error correction? See e.g. http://en.wikipedia.org/wiki/Error_detection_and_correction#Deep-space_telecommunications . I suspect that the 1e-12 rate you quoted for ethernet is probably the rate of errors that sneak by the built-in error correction. If 1e-5 is the error rate before error correction then the error rate after error correction would be too close to zero to worry about (with appropriate choice of codes). If 1e-5 is the rate of undetected errors after error correction then I would conclude that SpaceX must be sorely lacking in talented electrical engineers and computer scientists. If 1e-5 is the rate of errors that are detected but not corrected (except by resending) that's probably tolerable in this application.

BERs are generally before any forward error correction. There's a formula that relates FEC to SNR depending on the scheme, but I forget what it is!

I'm sure the numbers are right, probably some low power, low SNR transmission scheme. I'm just a wireline comms guy, hence the the 1e-12 numbers that I'm used to. IIRC, Gig Ethernet uses FEC but 100MBps and lower didn't. HDMI "sort of" uses it around non-video data. Wireless is different again, and completely NOT my area of expertise.

[corrodedNut]

This is an actual update:

http://www.spacex.com/updates.php

[apace]

This is an actual update:
http://www.spacex.com/updates.php

Looks large enough for crew also... like to see the mockups for crew. Wasn't there a milestone this months about?

[Space Pete]

This is an actual update:
http://www.spacex.com/updates.php

Looks large enough for crew also... like to see the mockups for crew. Wasn't there a milestone this months about?

Interesting - looks like the CBM is heavily clad in white panelling - probably heat shielding.

[Chris Bergin]

This is an actual update:

http://www.spacex.com/updates.php

A couple of nice photos, basically. The rest is aimed at people who are brand new to space flight.

Nothing in there the media could use, other than a photo or two. More aiming at the twitterverse, and more avoiding the media.

@HappyClapper: OMG, look it's a spaceship. Wow, thanks SpaceX you rock!!