LibreVNASniffer

Paper: https://www.mdpi.com/1424-8220/25/15/4654

This project is a software package designed to work with the LibreVNA and the Measurement Computing USB-2001-TC using a Type T thermocouple.

There are a few entry points into the project, which are listed below:

---

Gui.py

This is the main window of the program. Running Gui.py allows the user to access several functionalities:

Live Recording of VNA Data

With a computer running LibreVNAGUI, and a connected LibreVNA and Measurement Computing USB-2001-TC, the user will arrive at this UI:

Buttons

• Analyze – Begins recording a touchstoneList with the provided parameters. If parameters are missing, an error will occur.
• Save Run – Saves the current touchstoneList to a CSV in /data/timestamp/timestamp-csv.csv.
• Pause Run – Halts the collection of touchstones until resumed.
• Open Saved CSV – Loads data from a previous recording instead of showing a figure of the currently collected data.
• Take Screenshot – Records a single Touchstone file instead of one every 10 seconds. Useful for capturing individual $S_{21}$ readings.

Notes

• A note section allows the user to enter run-specific notes.

Sweep Configuration

• freqStart
• freqEnd
• points
• signalName – Must match a signal name within LibreVNAGUI.
• maxDB – Sets the upper y-axis limit of the figures.
• minDB – Sets the lower y-axis limit of the figures.
• IFBW
• bufferSize – Deprecated; does nothing.
• distance – Distance from the sensor in mm.

Console Output

• Error messages will appear here.

---

bayesianOpt.py

Entry point for Bayesian optimization of model parameters.

---

candleStickRefactor.py

Helper Python file responsible for generating summary figures from a collection of datasets. Some of these figures include:

• Resonant Frequency vs Distance
• Accuracy vs Distance
• Initial Frequency vs Distance
• $R^2$ Scores vs Dataset

When this script is run, all data in /data/ is used, and the figures are saved in /figures/timestamp/*.png.

---

csvList.py

Helper file that contains the final 28 datasets in a list called sensVsDist.

---

hysteresis.py

Responsible for plotting the hysteresis-over-time figure.

---

modeltest.py

This file runs tests on a selection of models with a set of datasets. It performs:

• Preprocessing
• Model instantiation
• Training
• Performance analysis

Model results are output as figures to /figures/timestamp/*.png. These include:

• Model learning curves
• Predicted Temperature vs Real Temperature plots

---

screenshotPlotter.py

Helper file responsible for creating the time-gating figures.

---