Plotting Signals



Here are a few simple utilities to convert raw datafiles into a format you can read and plot w/ gnuplot.

raw2num.c - converts a file of 16 bit shorts into a numerical human readable file with an index. Used for files created with

   gr.file_sink (gr.sizeof_short, "datafile")

Usage $./raw2num <input_file> <output_file> <offset> <number_of_numbers>

raw2num_f.c - converts a file of 4-byte floating point numbers into a numerical form with an index.

raw2num_fc.c - converts a file of 4-byte float complex pairs into numerical form.

sign2unsigned.c - converts a signed short file, numbers in the range of -32768 to +32767, to unsigned 0-65535, with the former 0 at 32768+bias. For example, my ADC creates numbers from 0-4095 with signal zero at 2048. When I create a signal file with gnuradio I can convert from it's signed format to unsigned with the 0 level at 2048.




Usage - start gnuplot. At the gnuplot> prompt type

gnuplot>plot "file"

will display the converted short or float file. Change the x-axis range with

gnuplot>set xrange [0:512]
gnuplot>plot "file"

For example, to use gnuplot like an oscilloscope for dial_tone.py, a version which adds both 350 & 440 hz into one channel
Briefly run $./dial_tone_2.py and hit enter to stop it. There will be a data file "dial_tone.dat" of float data. Convert to numerical with

$./raw2num_f dial_tone.dat dial_tone.num 0 65536

then display with gnuplot.

$gnuplot
gnuplot>plot "dial_tone.num"

and get a display like this:



Finally focus on a small area of the waveform with

gnuplot>set xrange [0:2000]
gnuplot>plot "dial_tone.num"

and get a plot like this:






For complex float output your have to run gnuplot with 'using'. For example, this program creates a complex signal

Run this briefly and get a complex float data file dial_tone_c.dat. Convert to numerical with

$./raw2num_fc dial_tone_c.dat dial_tone_c.num 0 65536

Then plot with

$gnuplot
gnuplot>set xrange [0:1000]
gnuplot>plot "dial_tone_c.num" using 1:2, "dial_tone_c.num" using 1:3

That creates one plot using columns 1 and 2, the imaginary part of the complex signal, and another plot using columns 1 and 3, the real part, so you get a plot like this:



Notice the 90 degree phase shift between the real and imaginary plots. Change the frequencies to negative in the above script, repeat and notice the change in phase to this