Week 1 (5/28 - 5/30)

We installed and familiarized ourselves with GNU Radio.

We also explored the architecture of the Orbit test bed.

Reviewed several papers to gain insights into the current state of 5G networks and the interference mitigation techniques used across different frequency ranges.

Week 2 (6/03 - 6/06)

We delved into the context of HyPhyLearn and conducted an in-depth exploration of Domain Adversarial Neural Networks (DANN).

Ran reference code for DANN and examined the source, target, and domain accuracies.

Analyzed graphs that demonstrated the model's ability to learn from domain-invariant features.

Week 3 (6/03 - 6/06)

We explored the TensorFlow implementation of the HyPhyLearn model, which classifies 2D Gaussian datasets.

Augmented the code from TensorFlow 1.0 to PyTorch and validated the experimentation.

Additionally, we explored how Dynamic Exclusion Zones apply to our mitigation objective. We weighed various approaches, debating whether the satellite receiver should autonomously localize interference sources or instead initiate a centralized algorithm for optimizing resource allocation.

Week 4 (6/17 - 6/20)

Emulated a DVB Satellite Transmitter in GNU Radio for the ORBIT Sandbox 1 - ran into some issues with existing implementations.

Formulated specific plans for our data collection experiments:

Experiment 1 (SAT Reception)

• SAT Transmitter
• SAT Receiver
• 4 sensors that compute the FFT of the received signal

Experiment 2 (5G Interference)

• 5G Terrestrial Transmitter
• SAT Receiver
• 4 sensors that compute the FFT of the received signal

Experiment 3 (Network Coexistence)

• SAT Transmitter
• SAT Receiver
• 5G Terrestrial Transmitter
• 4 sensors that compute the FFT of the received signal

Designed the neural network's input/output formatting.
\

Week 5 (6/24 - 6/27)

This week, we made significant progress in developing and validating our SNR Sensor. The sensor effectively differentiates the original signal from noise, allowing for accurate SNR value calculations. To ensure the sensor's accuracy, we also used FOSPHOR to verify its performance, which alleviated our initial skepticism about the readings. Now we have to test it for our 3 experiments.

We successfully implemented a Satellite Transmitter (DVBT) and a Satellite Receiver in GNU Radio. For the receiver, we developed and integrated a new block that converts stream data into vectors, uploading them as text files within the node. These files serve as collected data to train our machine learning model to predict SINR values, which are our target labels.

During our first experiment, we encountered an issue where the connected UHD devices did not correspond to the physical location of the nodes. To address this, we are planning to connect to nodes using SDRs at different grid locations.

We also working in a stage for automating the entire process to eliminate the need for the GNU Radio GUI. The following steps were taken:

• Extracted Python files and parameterized them, varying the transmitter gain.
• Created functions to log into different nodes.
• Streamlined the emulation process, including running time and data storage.
• Conducted one experiment every three seconds.

This automation would enhance our efficiency and consistency in data collection and analysis in the future.