search

erikkaashoek / tinySA

tiny Spectrum Analyzer
225 stars 50 forks source link

v1.3-5x / Can't capture short RF data bursts #48

Open rleto577 opened 1 year ago

rleto577 commented 1 year ago

I've been using v1.3-429 to capture random ~8ms RF bursts of data on the 315MHz ISM band and I noticed that I'm no longer able to capture these data bursts with v1.3-5x. What has changed to prevent me from capturing short bursts of data with v1.3-5x ? I've tried versions: v1.3-504, v1.3-506 and v1.3-513 without success.

Thank you in advance for your support,

Regards, Rej. VE3MPV

erikkaashoek commented 1 year ago

I tried to replicate using a 1000 cycle burts at 30MHz every 1s and this works fine. Please explain how to replicate

rleto577 commented 1 year ago

Thank you for the update. TPMS_BW

I upgraded the firmware to the latest version and I'm able to capture data bursts again. However, the frequency and amplitude vary over time as shown in the attached photo. I tested the burst with SATSAGEN and Rational Waves software and the frequency and amplitude remain very simular. I don't know why the TinySA exhibits these changes ?

erikkaashoek commented 1 year ago

If these are short bursts the position where you detect the burts will depend both on the frequency being sweeped and the moment of the burst You have a 87ms scan time and a burst of 8ms, so a burst will take 9% of the width of the sweep (hence the 40kHz of the yellow trace as the sweep is 500khz wide). The overall shape (green trace) is caused by the RWB being 100kHz (-3dB width) and the burst being send at 314.98MHz The FScan sweeps very fast and causes to miss some power as the time to load the power detector is on the edge Check if normal scanning remove the "holes"

rleto577 commented 1 year ago

Thanks for the explanation. As shown in the attached saved screen image, the maxHold trace indicates over time that the frequency and amplitude vary significantly causing reading errors. Based on your explanation, this behavior appears to be normal. As you suggested, could kindly you include the configuration for a normal scanning and what do you mean by "holes" ?

Thank you I advance.

erikkaashoek commented 1 year ago

Why don't you use zero span mode?

rleto577 commented 1 year ago

Thanks for the suggestion, however I'm interested in finding the mid frequency of a pulse. Considering that the center frequency is unknown, It defeats the purpose of the zero span function which is primarily used for signal demodulation, channel power measurements and measuring envelope of a pulse at a known frequency.