[DSP-10] Here it is, the single antenna, brickette-only EME2 detection.

[Courtney Duncan]

>Hi Courtney,
>
>Fabulous work!  This is indeed going towards the edges of 
>weak-signal work!  I really like your enthusiasm, or is it passion, 
>for this capability to dig for signals without limit!  Like EVE!!
>
>One additional thought, or maybe one that you have already explored. 
>Detection theory sets a threshold above the average noise power to 
>allow a prescribed false alarm probability. The Central Limit 
>Theorem says that the statistics of all long-term sums are Gaussian, 
>so setting this threshold is done with the cumulative Gaussian 
>tables.   A threshold at your 4.2 sigma point says that the 
>"probability that is was NOT an EME echo" is 1 in 50,000.  Sure 
>sounds like EME to me!!

Also, if it were just luck, the best peak could well be in any of the 
other bins, and might just as easily be "down" as well as "up". 
Doesn't that mean another 41:1 against this being mere "luck"?

(I had looked up the error function for 4.2 sigma in my 1968 CRC.  It 
only goes to four places so it said 0.5000, for one tail, so better 
than four and a half nines for both which agrees with 50,000:1.)

>On the 2-meter noise levels, be sure to include the minor lobe Earth 
>and Sun as well as galactic contributions.  An interesting test is 
>to switch between the antenna and a load to measure this.

I plan to measure the noise figure of the box itself then use that to 
infer noise temperatures in various data sets past and future.  I'll 
probably want to revise the results when I have that information.

>Keep up the great work!

Thanks for the encouragement.  This radio and the approach it 
represents is a great tool with potential that we haven't thought of 
yet!

Courtney, n5bf/6

>
>73,
>Bob W7PUA
>
>
>At 12:43 AM 6/21/2006 -0700, Courtney Duncan wrote:
>>http://home.earthlink.net/%7En5bf/eme2.html
>>
>>It's a fairly long story!
>>_______________________________________________
>>DSP-10 mailing list
>>DSP-10 at mailman.qth.net
>>http://mailman.qth.net/mailman/listinfo/dsp-10
>>
>
>
>_______________________________________________
>DSP-10 mailing list
>DSP-10 at mailman.qth.net
>http://mailman.qth.net/mailman/listinfo/dsp-10