RFSoC_SDR

Steps to bring up an RFSoC4x2

• A very good source of information is provided in This Link and This Link

• Download the RFSoC4x2 image from here. This repo is tested with v3.0.1 but newer versions should be also OK unless they are not backward compatible.

• Program the RFSoC4x2 image to the provided SD-card usign Rufus (in windows) or any other similar softwares in your specific OS.

• Put the SD-card in RFSoC4x2 and power it on. Wait until you see the IP address information shown on the LCD.

• Connect the RFSoC4x2 to your laptop/PC using a USB cable.

• Open the RFSoC web interface using a web browser on This Link

• Put all the python scripts from here in the corresponding project folder on the board. you can create a folder for your project at /home/xilinx/jupyter_notebooks/YOURFOLDER/. You only need to copy the .py scripts not all other folders in the provided link.

• In the backend.py script, change import_pynq to True. Also if you're planning to use the Sivers antenna, please change import_sivers to True.

• Put the clock configurations files from this folder in the Repo in this folder on RFSoC4x2: /usr/local/share/pynq-venv/lib/python3.10/site-packages/xrfclk/

• Provide internet connection to the board by connecting it to a server/PC and routing the traffic or by any other methods you like.

• Install the RFSoC-MTS package on the RFSoC4x2 according to instructions in this linke

• Put the latest version of the RFSoC4x2 FPGA image files from here in your project folder beside the python scritps. You only need to transfer .bit and .hwh files. Please don't transfer the .xsa along with the other two files because it causes a kind of conflict in loading the image.

Steps to prepare the host computer for doing experiments

• First, clone the repository in your host computer.

• Then, we create a virtual environment. The command below will create an environment named env, but any other environment name can be used.

  python -m venv env

  The command may take several minutes, and it may not indicate its progress. After completion, the virtual environment files will be in a directory env. This directory may be large.

• Activate the virtual environment:

  .\env\Scripts\Activate.ps1  [Windows]
  source active env [MAC/Linux]
      or source bin/activate

  On Windows Powershell, you may get the error message “...Activate.ps1 is not digitally signed. The script will not execute on the system.” In this case, you will want to run:

  Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass

• The first time you activate the environment, install the package requirements:

  (env) pip install -r requirements.txt

• You can exit the virtual environment with:

  (env) deactivate

To use the package later, you will need to activate the virtual environment and run any commands in that environment.

Creating a requirements file

If you update the installation in the package, you may need to re-create the requirements.txt file with:

   python -m pip freeze > requirements.txt

If you do this on Windows, you should edit the file requirements.txt as follows:

• In requirements.txt, you may have a line like:

  pywin32==306
  

  Delete this line since it is only needed for Windows.

Steps to do measurements on FR3 using Pi-Radio FR3 Transceiver

• Assemble Vivaldi antennas using appropriate spacers to tune the antenna spacing needed for the target frequency.
• Install Vivaldi antennas on a fixed structure.
• Connect the RF ports of the Pi-Radio FR3 Transceiver to the Vivaldi antennas. Please be consistent in antennas port indexing. No passive elements are needed on the RF port if you're connecting them to antennas. But if you're using a cable to connect RF ports you need at least 20dB attenuator.
• Connect the IF ports of the Pi-Radio to the RFSoC4x2. The RFSoC4x2 outout ports 1,2 are DAC_A, DAC_B respectively. Also its input ports 1,2 are ADC_B and ADC_D respectively. Please connect them to the right port numbers on the Pi-Radio. On the transmitter side, connect a DC blocker and a DC-1300 MHz filter on the RFSoC4x2 ports. On the receive side connect a DC blocker, a DC-1300 MHz filter and a 20dB attenuator on each port.
• Power on the Pi-Radio FR3 Transceiver.
• Connect to the Pi-Radio board by running ssh [email protected] in the Linux shell. The password is temppwd
• run ./do_everything.sh to configure the board on the right frequency. At this point the Pi-Radio FR3 Transceiver is ready to use.
• Alternatively you can connect to the board's Web GUI at This Link and configure all needed parameters.
• Connect the RFSoC4x2 to your laptop/PC using a USB cable.
• Open the RFSoC web interface using a web browser on This Link
• Run the rfsoc_test.py script on the RFSoC4x2 using a jupyter notebook or using python rfsoc_test.py. When you see Waiting for a connection in the log, the RFSoC4x2 is ready.
• refine the configurations in the rfsoc_test.py file on your system/PC and run the script using python rfsoc_test.py. You need to run the code in an envioronment with capability of showing plots which means the backednd to show the matplotlib figures should be cofigured propoerly. I suggest you use VScode as your IDE.
• First time you run the system you need to do phase calibration on the receiver, so when the script shows a message about the calibration, follow the instructions to run the calibration.
• Finally the script should receive the measurements, do the needed processing and show the animation plots.
• You can also save the signal and channel responses by adding appropriate elements to the save_list parameter.