Setup for GSM/EDGE Uplink (high crest factor) measurement
I am trying to measure a GSM uplink signal in the 900MHz band. The measured signals are not stable as in each sweep the power level is always very different (seems to jump after each sweep). What can i do to avoid this? What is the best setting to capture a uplink signal in the 900MHz band?
The GSM uplink offers a high crest factor so if you want a "flat" noise level" its important to get the dynamic working. After generating a test file with our SMJ100A (Uplink with single framed and -30dBm) i would suppose to deactivate the internal preamp and add the internal attenuator (10dB should work fine).
If you are measuring with a standard 120kHz RBW you should use a long sample time (5000mS or higher) to capture the GSM signal (If you sweep the full GSM900 frequency band with a span of 39MHz). With 5000mS the measured power level will be 10dB to low but you will capture and see the signal. Rising the sample time will get you to the right power level. Attached is a picture of my setup showing all settings with 5000mS. This is working fine to capture the signal.
After zooming in the signal (1,6MHz span) you can reduce the sample time to 2000mS if you use 1000 or more samples and the power level will be perfect too. See picture attached.
Also see viewtopic.php?f=31&t=872