I am testing the AD9371 Rx path using the ADRV9371-W/PCBZ EVB.
I am using the Xilinx, HW-Z7-ZC706 Rev: B evaluation platform and the TES software to control the EVB.
Rx Profile: 100MHz, IQRate=153.6MHz, Dec5
The Rx PLL frequency is 3410MHz. The input signal is a CW signal at 3415MHz Pin = -30dBm connected directly to Rx2 SMA connector.
Rx2 gain Index = 241, It means that the AD9371 Rx Gain = 10dB in this frequency.
In file TES_Rx2_Plot_Index=241_Pin=-30dBm.JPEG we can see the TES plot of this signal.
When I capture the IQ samples to Matlab and perform FFT I get the following results as shown in file Matlab_Rx2_Plot_Index=241_Pin=-30dBm.JPEG
1. The measured signal in Matlab is -19.77dBm, it relate to the 10.23dB gain of the Rx path.
2. At low power levels (No RF, Noise test) the Matlab plot is identical to the TES Rx plot.
Files: TES_Rx2_Plot_Index=241_Pin=NO_RF.JPEG and Matlab_Rx2_Plot_Index=241_Pin=NO_RF.JPEG
My questions are:
1. Why the IQ samples in the Matlab plot looks like the Rx path is in compression ?
2. Why there is a difference between the TES Rx plot to the IQ samples Matlab plot ?
3. Why I don't see this difference in the noise test ?