search

keenerd / rtl-sdr-misc

A bucket of various work-in-progress rtl-sdr ideas.
213 stars 175 forks source link

How to make the heatmap larger? And see the frequencies #13

Open stillnet opened 8 years ago

stillnet commented 8 years ago

I'm a newbie to SDR, this might be a dumb question. I am having trouble figuring out how to make the heatmap larger, and show the frequencies at the top. Or at least somehow more readable. I have a csv file of 3Mhz of spectrum, with 5Khz steps. I recorded for about 1 hour. I'll attach the both the csv and png files here. You can see there are no headings (probably not enough room to label all the steps?). I'd like to make the whole image larger, so I can see more detail. Like 3x wide, and if I could make it taller that would help me see the short signals also. I've tried using --ytick but that doesn't seem to help. I'm not sure what I'm supposed to be giving ytick: seconds? milliseconds? pixels?

These were my arguments: rtl_power -f 145M:148M:5k 2m.csv

Thanks for any help.

2m

2m.zip

iDoka commented 8 years ago

I have been similar issue:

$ rtl_power -f 88M:108M:125k fm_stations.csv
$ ./heatmap.py fm_stations.csv fm_stations.png
$ file fm_stations.png 
fm_stations.png: PNG image data, 257 x 86, 8-bit/color RGB, non-interlaced

why I always get 257 x 86 image size?? how to I can control generated image size?

MeFisto94 commented 5 years ago

You guys have to adjust the -i parameter of rtl_power. For instance in the examples, -i 100 was specified. This means that every 100s the datapoints are written into the csv (which probably means the intensity is sampled over 100 seconds).

Thus the heatmap will have a height of sample_time/100s. It makes no sense to duplicate entries, you'd have to zoom in for that. Or specify a faster "-i" to capture more "dynamic" data. Be aware though that insanely small values decrease the quality of your capture again, because averaging the signal over a few seconds makes sense.