Message boards :
SETI@home Science :
What tests are being run?
Message board moderation
Author | Message |
---|---|
Tom Mazanec Send message Joined: 19 Aug 15 Posts: 79 Credit: 6,938,247 RAC: 4 |
I see the tests being flashed on the display as they are run, like "Pulse triplets" or "fast Fourier Transforms" or "Gaussian...something or other" But they are too fast for me to make out. Can you list the tests and what they are doing? |
rob smith Send message Joined: 7 Mar 03 Posts: 22535 Credit: 416,307,556 RAC: 380 |
Fast Fourier Transform is a technique for moving data from the time domain into the frequency domain. The data coming from the telescope is in the form of energy (signal strength) plotted against time. It is easier to analyse such data as energy plotted against frequency - hence doing the transform. The rest are looking at data in the frequency domain: Triplet detection looks for a series of three peaks at equal frequency intervals. Spikes are short, very narrow peaks Gaussian signals are signals that are "bell shaped", having a Gaussian distribution about the centre frequency. Doublets are similar to triplets, but only two such signals. Each of these signal types may be indicative of a "purposeful" transmission. BUT most actually come from our own endevours when further analysed. Then there is "auto-corelation", which is a sort of combination of Fast Fourier Transform and peak-shape detection. A bit hard to explain without several pages of maths, but can be very good at detecting potential data carrying signals that are too small for other techniques. One day soon I expect to see another transform technique being deployed - "wavelets" - this technique can be faster and more sensitive than FFT, but one has to be VERY careful in selecting the "seed" used as the seed can have a significant impact on the outcome. Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Mr. Kevvy Send message Joined: 15 May 99 Posts: 3806 Credit: 1,114,826,392 RAC: 3,319 |
Simple (I hope) summary of autocorrelation: The more information per time a signal carries, the more random-looking it becomes. Noise is a random signal, so the signal starts looking more like noise and thus can't be distinguished from the noise background. Autocorrelation consists of repeating each block of the transmission at the sender end, then at the receiver end each repetition can be compared to the last one which will isolate the signal from the background noise. Hope I got that right. :^) |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.