Discussion Forums

simple signal picture viewer program

3 replies [Last post]
Joined: 2010-08-27
Posts: 65

Here is the really simple signal picture viewer program.
It should work on anything that can run wxruby but I couldn't get it to work with linux.
It is just a ruby script so you can edit it and wxruby is pretty easy to understand.
The reason the file is 60 megs is because of all the pictures.
You need ruby and it seems like it will only work on windows.

I made the program because it was a pain to try to line up the pictures or make a spreadsheet to keep track of which ones were close to each other. I used wxruby because it is really easy to program. The down side to wxruby seems to be that it has some issues with flickering which I couldn't figure out and I couldn't get it working on linux. But hopefully that is because I did something stupid.

The found folder has all picture data separated into folders by data set. Don't put anything else in those folders or it might break. You can edit the cf.txt if you find the center freq is off.

If you want to add a picture make sure its in this format
and within that range.
to figure out what number you need for your signal you do something like this
(signal_freq - ((center_freq)-(8738132/2)))/1000 = picture_number

If someone figures out how to get it to work on linux I would like to know.

DANGER ignorant waters ahead:

To me something like this seems really useful but I am pretty ignorant when it comes to understanding most setiquest stuff. There is probably a good reason why this is not a useful approach to finding signals that have potential. The only thing that I can think of is the data I am using is no where near real time. What is the fastest you could get the data from the telescope to the wiki data set format? I am pretty sure I could make the picture generator I use a lot faster and then get some friends to help me go through a data set. So it takes me about 40 mins to go through one alone. If I manage to bug them enough and got 5 people it would only take 8 mins. Then you could have the program show you where the signals are and which ones are obvious RFI pretty quick and still be able to target anything interesting before it goes over the horizon. Could the telescope ever be that responsive to where you can look one place for a couple of minutes then look at another place then another and have people visually going through the data sets trying to keep up with the telescope and then point the telescope back at anything interesting people find visually before it goes out of range? I have gone through a lot of data sets and it seems like most of it is RFI like people have said. I know sonata does some magic I will never understand but the dots seem to be a good example of something that more data should have been collected on unless sonata ruled it out somehow with the voodoo it does. If you had people looking maybe they would have noticed it before it went out of range and you could have collected more data to figure out what it was. Maybe you could set this all up once a month and call it Signal Madness day.

One thing I have wondered about is lets say you see one signal that is interesting but really faint. Can you focus the telescope in on that frequency to make it less faint if you know what frequency to look at?
END ignorant waters:

Joined: 2010-08-27
Posts: 65
Use this link to download

Use this link to download instead of the other one:


this one has an extra data set and now you can shift + right or shift + left  arrow to jump 100000 hz

my plan is to try to go through all the 1420mhz data sets and see what happens.  

Joined: 2010-08-27
Posts: 65
here is a exe version if you

here is a exe version if you don't want to install ruby.  Windows only though.

Joined: 2010-08-27
Posts: 65


At the bottom of that page I will try to post links to the signals I see then you can download them and unzip them to your found directory to get the program to show them.  Just with the ones I have gone through so far it seems like it is really easy to spot RFI.  

make sure you get the directory structure the same as the ones that are already in there if that makes sense.

something like
/found/dataset-name/then pictures

not like this
/found/dataset-name/dataset-name/then pictures

which seems to happen when people unzip things.