SNR


#1

I’m testing various antennas. It would be neat if someone would come up with a way to measure average signal to noise ratio lets say over a half hour period. It would be a command like sudo piaware-snr


#2

This might be a little more involved than you want, and may be not exactly what you are looking for, but if it is, we can help. Check out http://discussions.flightaware.com/post178511.html#p178511 and see if that is along the lines of what you are thinking of?


#3

This might be a little more involved than you want

Wow I see 6 pages. Lots of work and contributions going on. I feel a bit naive. My idea would just be a number returned saying this antenna is better than that antenna. I suppose I could just watch the number of planes I see, but that changes constantly. I’m in LA area, so LAX is pretty busy.


#4

I know it seems to be a bit intimidating to some at first, but once you have the graphing set up, you can get a lot of information. (And there are a lot of people that are willing to help you out with setting things up.) It’s kind of rewarding to see how everything works, there is another thread where abcd is using the graphs to test out antenna settings at http://discussions.flightaware.com/post178675.html#p178675.

I’m right under the final approach glide path for KSNA, right about EFFIE, there are a few of us SoCal users here in the forums.


#5

it seems to be a bit intimidating

Maybe, I should start at page1? However it might be tough to follow the back and forth exchanges. I’m not really a newbie, but in this area, I wouldn’t know enough to ask an intelligent question right now. Are all the graphs out of a hardware spectrum analyzer or from a software package I can run on a PC?

I’m in north part of SF Valley, near Granada Hills. So quite a ways from LAX, but North of KVNY airport and KBUR Southeast.


#6

Good idea to start at page 1, but read through all the pages, there are some changes in the way the software is set up the farther you go in. (If you use the install instructions from page 1 you might end up having to reverse and redo things when you get to pages 2 and 3 for example.)

The graphs are all made directly on the Pi, no other hardware is needed. The main program is called ‘collectd’ and it is set up to create databases called RRD’s (Round Robin Databases.) That daemon is configured with a config file that needs some instruction about what information you’d like to collect and where to collect it. It will need some files from dump1090, a lot of the extra graphs are from a version called dump1090-mutability that is written and maintained by a user you might see around a bit, obj. There are also instructions on how to install dump1090 using the mutability fork if you need some pointers with that.

Once you have the data in the RRD’s, the program that does the conversion from data to graph is the RRDTool. That also runs on the Pi and you’ll find out how to control that with the make-graphs.sh that people have customized and configured. It runs on a regular schedule, controlled by a cron job. Again, ask for help if you’re stuck. It’s nearly impossible to break things and if things go awry you can always start over again with a new SD image and try.

Finally, after you have the graphs made you need to be able to see the images of the graphs, and that’s done with a web server. The recommended one is lighttpd as that will let you display both the plane map as well as your newly made graphs.

So, read through a few times, see how and what people have done and then jump on it and give it a go, I think you’ll find you’re more capable with this than you may think.


#7

I am actually working on something similar to this. It is a python script that connects to multiple instances of dump1090 and gathers the reported RSSI from certain messages.


#8

If you go down the collectd route, once you have the raw data it’s fairly easy to generate a graph that shows, say, the message rate ratio between two receivers using rrdgraph’s little RPN expression language.


#9

once you have the raw data

Thanks all. As I read, I’m beginning to understand why more effort has been done. The graphs provided by Flightaware give one a start, I can see getting the data to help improve the setup, but showing it different ways helps to teach relationships between the variables. Plus this topic is a good example of sharing by those who know to those who don’t. Thanks again.