AFF3CT comes with a Bit/Frame Error Rate (BER and FER) specific visualization tool: the PyBER plotter. PyBER is written in Python and directly plots curves from the standard AFF3CT output.
First install Python 3 and the Python 3 package manager PIP:
$ sudo apt-get install python3 python3-pip
Next, install the required dependencies to run PyBER:
$ sudo pip numpy pyqt5
$ python3 pyBER.py
By default, PyBER shows the folders and files located in the
tests directory. The program only shows the file with the following extensions:
tests/data folder contains a set of simulations pre-ran with AFF3CT. You can select the corresponding files in the
data folder and take a look on the decoding performances. You can display more than one curve at a time by selecting a set of files (keep the
maj key pressed and select the files you want to compare with the left mousse button).
Lets assume that you have compiled AFF3CT and you are in the
build directory. You can launch a new simulation and redirect the standard output in a file like this:
$ ./bin/aff3ct --sim-type BFER --cde-type TURBO -K 1024 -N 3072 -m 0.0 -M 1.01 --dec-type LTE --dec-implem FAST -i 6 > ../tests/data/turbo_simu.txt
You have just created a new file (
../tests/data/turbo_simu.txt) and you can visualize it. Notice than PyBER is able to update the curves in real time (no user action is needed to refresh the selected file(s)).