search

junzis / pyModeS

Python decoder for Mode S and ADS-B signals
GNU General Public License v3.0
546 stars 153 forks source link

implementation of uncertainty values from csv files #24

Closed JoseAndresMR closed 6 years ago

JoseAndresMR commented 6 years ago

Hello @junzis ,

Some tips about the code.

Pandas library is directly uploaded, I suppose you are using it from your common files, so you should change it. About csv path locations, they are on pyModeS/decoder/adsb_ua_parameters. You should move them as well as change all paths on the code. This brings another issue: we shouldn't upload the csv every time a msg is recieved.

By the way, I'm sorry for all bugs on my last pull-request, I would have avoided it if I had a way to check, like the pmslive you showed to me on the issue. How can I access that? Is it just pyModeS or I should dive in BlueSky (which I'm going to do anyway cause it is also very interesting).

Thank you in advance for your help.

junzis commented 6 years ago

Looks great! let me checkout the code and test with our receiver. No problem for the bugs in the previous pull. It's easy to fix.

We have a receiver setup on the top of the building in TU Delft. Unfortunately, one has to be on the campus network to access the TCP stream. But if you have RTL receiver, it can be set up easily. Here is a library I wrote to work with these receivers: https://github.com/junzis/sil

JoseAndresMR commented 6 years ago

Thank you! We don't have a RTL receiver yet but on the search. Meanwhile, it seems suitable the RaspberryPi-based of FlightRadar24, which offers as well the use of their global stream. As soon as we receive it, I will get on /sil.

junzis commented 6 years ago

Yes, the FR24 RaspberryPi is indeed connected to an RTL receiver, it is a good gadget to have.

junzis commented 6 years ago

in fact, the stream modules that were developed in sil is integrated into pyModeS now. So, in principle, you can connect pyModeS directly to a beast or avr raw stream. To do so, you just have to use a tool, such as ModeSMixer, to read the RTL and convert it to a raw stream using TCP.

junzis commented 6 years ago

Hi @JoseAndresMR, I made some update to the code, and merged the pull request.

Now the uncertainty data is coded as python variables. So we do not need pandas or numpy to load the data. This is much fast for execution than load the files each time the functions are called.

I also re-wrote some of the functions, especially those related to NIC. Now they should be easy to read and debug.

JoseAndresMR commented 6 years ago

Hi @junzis,

We are definitely going to try it given your advice. I can't wait to use our own stream. In fact, i'm going to use it on my repository as well as pyModeS.

About merge, I'm happy for the news. I have a far way to improve my coding because I'm quite new on this and is even my first time colaborating on GitHub. But actually learning a lot, so thank you so much.

Finally, if you find out something else I would be useful for, don't hesitate to tell me.

junzis commented 6 years ago

@JoseAndresMR, nice code already. What I did were just optimizations. cheers!!