SETI Signal Type Distribution


log in

Advanced search

Message boards : SETI@home Science : SETI Signal Type Distribution

Author Message
Bill ButlerProject donor
Avatar
Send message
Joined: 26 Aug 03
Posts: 54
Credit: 1,114,122
RAC: 1,657
United States
Message 1370144 - Posted: 20 May 2013, 22:54:55 UTC

On the Science status page, the Master Science Database table - I wondered what the distribution of those huge numbers was. Easy pie chart. Here are the results.






____________

Profile Bob DeWoody
Avatar
Send message
Joined: 9 May 10
Posts: 1621
Credit: 575,165
RAC: 347
United States
Message 1370190 - Posted: 21 May 2013, 4:06:07 UTC

So, what is the significance, if any, of the distribution?
____________
Bob DeWoody

My motto: Never do today what you can put off until tomorrow as it may not be required. This no longer applies in light of current events.

Bill ButlerProject donor
Avatar
Send message
Joined: 26 Aug 03
Posts: 54
Credit: 1,114,122
RAC: 1,657
United States
Message 1370339 - Posted: 21 May 2013, 20:31:36 UTC

That is likely a question for the ongoing science research.

What I noticed:


    1. The short term 24 hour observation doesn't differ much from the entire signal database. The distribution of signal types is about the same. I was a little

    disappointed by that. I was hoping the sky would be more robust. One expects the huge longterm database to average out signal distribution. I wanted the short term

    distribution (24 hrs) to show a little "noise" and change significantly by comparison. Not sooooo.


    2. About half of the signal database is spikes, a suprise to me. This might emphasize the value of Astropulse. Astropulse searches for brief (0.4 μs to 204.8 μs)

    radio pulses. I guess I can call them spikes.


    3. I was pleasantly suprised by how many triplets are in the signal mix. We think these are on the high road to assembling an ETI signal.


    4. Caveats:

    We don't know how the signal distibution is affected by purely technical factors like:

    Arecibo just looks at a narrow band across the sky. Maybe other radio observatories will see a different signal distribution across the broader sky.

    I don't know what project scientists think about the amount of rfi contamination in the signal database. Hopefully the database is reasonably clean since there are rfi countermeasures in hardware and software.

    Potentially, the signal distribution could change if irrelevant technical side effects are removed or a broader SETI database is built.



____________

rob smithProject donor
Volunteer tester
Send message
Joined: 7 Mar 03
Posts: 8532
Credit: 59,211,879
RAC: 82,277
United Kingdom
Message 1370349 - Posted: 21 May 2013, 20:43:13 UTC

A lot of effort is given to the problem of RFI, and blanking it out.

The orientation of Arecibo is such that its main axis is pretty much along our arm of the Milky Way, so it probably covers the bulk of the sky with the highest potential. However, as you say, other observatories will give a different view, particularly into the "off plane" regions.
____________
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

Josef W. SegurProject donor
Volunteer developer
Volunteer tester
Send message
Joined: 30 Oct 99
Posts: 4301
Credit: 1,068,713
RAC: 977
United States
Message 1370351 - Posted: 21 May 2013, 20:44:24 UTC - in response to Message 1370190.

So, what is the significance, if any, of the distribution?

Well, Spike searches are done first, so when there's a lot of RFI tasks tend to overflow and add another 31 Spikes to the total. And because Gaussian fitting is only used on WUs with medium angle range, the total for those is lower.

In short, the distribution indicates the thresholds for reporting signals are tuned as intended. That's about a 50% chance of normal noise being seen as one reportable signal in each WU.
Joe

Profile ML1
Volunteer tester
Send message
Joined: 25 Nov 01
Posts: 8490
Credit: 4,190,886
RAC: 1,809
United Kingdom
Message 1370403 - Posted: 21 May 2013, 23:35:20 UTC - in response to Message 1370351.
Last modified: 21 May 2013, 23:36:27 UTC

So... That reported distribution is most likely an artifact of the analysis algorithm...


I find it interesting that even for what is essentially random input data, there are over 20% of the entries for triplets found. Also very intersting is that a 24hour period is very similar in results to the whole average over the many years.

Ofcourse, of far far greater interest is whether there are any few consistent sources with consistent findings for any sort of non-natural signal!


Keep searchin',
Martin
____________
See new freedom: Mageia4
Linux Voice See & try out your OS Freedom!
The Future is what We make IT (GPLv3)

Message boards : SETI@home Science : SETI Signal Type Distribution

Copyright © 2014 University of California