"Martin 53N 1W" <ml_news@ddnospamddml1dd.co.uk.dd> wrote in message
news:h3M%c.550$E56.47@newsfe4-gui.ntli.net...
<snipsnap>
First of all, I think we should forget about finding beacon signals
(narrowband, deliberately
aimed signals with a message for us) . 1240 Mhz is way to low a
frequency for interstellar beacons
(too expensive to operate for ET).
Mmmm... That cuts out the most /easily/ detectable stuff for what
detection capability we have available at present.
Yes, sorry. Nobody said it would be easy to find proof of ETI.. :)
I believe that it is much more likely that ET signals out there
detectable with Arecibo - seti@home,
will be CW radar signals (semi-long (10-100sec), narrowband radar
pulses), and these
will be the first ones we will pick up.
Whether the Arecibo + s@h (classic) system could detect a
seti-equivalent our own radars is a good question... and how far out?
(Wooley estimations please? (:-))
David has actually talked many times about this, and seti FAQs also has a
lot of data.
Also, Alfred is attempting to set up a detailed 'radar' profile of earth,
which should provide more accurate data.
Very ball-park estimates, I think seti-equivalent of our own radars
detected with an Arecibo receiver, varies from a few AU's
(for short-distance, wideband, surveilance radar), to
100's to 1000's of LYs for high-resolution military radar and
Arecibo's asteroid radar in CW mode.
That is, if you listen on the right frequency....
And recognize the signature of a long-distance, high-resolution radar
signal...
I'd think that would be a narrowband, CW signal, received here
as a prolonged 'pulse' of 10 to 100 sec (received as a Gaussian),
possibly linearly shifting in frequency (to enhance distance and
velocity readings of the radar).
Very much like SHGb02+14a........
That's why I think we should not discard such signals...
Unless the source solar system is doing a LOT of radar, I suspect we're
sunk without trace at source (:-((
Good point.
Still, looking at our own history, we have send of many millions of times
more high-power, narrowband CW radar than we have send out deliberate
beacon signals....
[...]
Here are some suggestions :
(1) To rule out RFI you need to recognize it : To automate recognition,
you need to
have a database of know RFI signals or some pattern analysis.
[...]
Yep, and there's various ways of doing this. One question is whether to
have a central database, or to have a distributed database with some
form of peer-to-peer between contributors. Also, how do you handle the
90% of users that try for a few hours and then get bored never to be
seen again? Worse, what of those users that appear 'reliable' and then
just vanish without trace? Data loss??
I'm not too concerned about data loss and vanishing users.
Whenever a complex task is outsourced to a seti@home - like distributed
system, a data integrety method needs to be put in place. seti@home
already has that (if you dont return a result for your work unit, it will be
send to someone else).
The hassle of this (distribution and integrity checking) is worth it if the
task
is complex enough.
For RFI analysis, or correlation analysis, I'm not sure if that is
computationally
so intensive that it requires a seti@home distributed computing approach.
But it sure opens up a lot of new analysis methods.
I just don't know enough about the RFI filtering and data analysis/scoring
mechanisms that seti@home already has in place to decide if it's a good
idea to outsource this to a new screensaver.
<snipsnap>
The correlation analysis done for the 200 interesting signals was +-75Hz
or so.
With a large computer farm, a much more thorough correlation can be done
on all same-sky-different-time work units.
A big problem there is practical access to all the WU results. The
database is likely too large (and controversial) for a raw download for
multiple public users.
If you have a _serious_ proposal backed up by well argued and well
researched methods, then you could try offering you proposal for
consideration...
I'd like to do that, because seti is definely a passion of mine, but I want
to know that I'm solving a problem that is relevant.
So I first need to understand the problems that are currently facing
seti@home.
But the problem I think is that there is just not enough scientific and
technical
info coming out of Werthimer's group. Not what the problems are in the final
signal selection process, not how it's done, not what is done, no plan of
what to do when data increases 10 fold, not anything.
The seti@home group seems to be always busy.
Also, where do I send inquiries or requests ? No email posted anywhere...
After we are done processing a work unit, information seems to disappear
until press releases come out.
For example, we still dont know all the details about this signal
(SHGb02+14a).
Bandwidth, duration, how closely it correlated with its repetitions,
how often it repeated, what the time duration was between repetions,
between observations, how often it was not detected (when observing this
sky location) etc etc.
And it gets worse : there is no indication that Werthimer is going to
re-observe this
(or any other high-scoring) signal in the future.....
I have a gut-feeling he will, but there is no message at all to that extend.
Seems that he was very much trying to stop the media from continuing
talking about this signal. This distorts scientific/technical data.
So we still dont know why this signal was discarded (when rated as
'promising' earlier),
or if it really was discarded or not, or if there are plans to adjust
scoring
so that different, possibly more interesting signals come up.
After all, it's pathetic that after 5 years of analyzing data, we dont
even have a Rio level 1 signal (insignificant) or level 2 (low) signal.
http://seti1.setileague.org/iaaseti/rioscale.htm.
Actually, Werthimer's conclusion that signal SHGb02+14a is
Rio level 0 (zero probability, no importance) of being of ET origin, is
really strange.
How can he know there is zero probablility ?
Does he know something we don't ?
Or does he just say this to silence the media, and it is actually a Level 2
or 3 signal.. ?
So, I think the real problem is that there is not enough info coming out
of Werthimer's group. Especially not about the final 'candidate' selection
and Rio - scale rating and selection process.
Need I continue ?
Many things are possible!
Yes. If we know which problem needs to be solved...
But I'm not convinced that problem is technical at this point...
Meanwhile, there is still the danger of an overenthusiastic ignorant
media shouting wolf a few too many times and then discrediting all the
seti projects.
As long as we keep the top-1000 list open and steady at 1000, there will
not be any media attention.. Just more seti enthausiasts sifting through
prospect signals.
It's when we say that one particular signal is 'promising' is when media
are interested.
The lack of a steady stream of signals claims/disclaims is partly what
has the media so desperate for /any/ claim. If there are no claims, then
the media people are very happy to make up their own hyped claims as has
recently been seen!
Actually, if you really read the New Scientist article, there is not that
much 'hype' in it.
http://www.newscientist.com/news/news.jsp?id=ns99996341.
Just a few quotes from our seti scientists.
And I doubt that these quotes are incorrect.
Do you ?
Perhaps we should have a monthly 'best hits' web page, similar to the
'Near Earth Objects' and other astro-search bulletin type pages. The
problem there is for someone at Berkeley to have the time and interest
to program an automated web page...
Actually, Berkeley already has a top-25 list :
http://setiathome.berkeley.edu/Candidates/index.html
Just that they do not provide the raw data (work units) with them.
And that would be so cool to have.....
Something that would be good would be to Boinc up some signal analysis
that takes advantage of the Gflops of nVidia GPUs!
eg:
http://www.bionicfx.com/media/BIONICFX_09012004.pdf
ftp://download.nvidia.com/developer/presentations/2004/SIGGRAPH/ImageProcess
ing.pdf
http://developer.nvidia.com/object/fast_math_routines.html
http://developer.nvidia.com/docs/IO/8229/Cg-Paper-Slides.pdf
(FFTs:)
http://developer.nvidia.com/docs/io/4106/Technology_Directions.pdf
http://developer.nvidia.com/object/sdk_samples.html
(Get hacking (:-))
Thanks for the links !
Keep searchin',
Martin
--
---------- OS? What's that?!
- Martin - To most people, "Operating System" is unknown & strange.
- 53N 1W - Mandrake 10.0.1 GNU Linux
---------- http://www.mandrakelinux.com/en-gb/concept.php3