Reference Number: ARL-SRP-00024
Project Description
This project investigates how deep learning can support real time radio frequency (RF) signal classification on resource constrained edge devices. As wireless environments grow more complex, quickly recognizing signal types is important for effective spectrum awareness. Traditional signal processing methods can be too slow or computationally intensive for small, low power platforms.
The objective of this internship project is to explore lightweight deep learning approaches that could eventually be deployed at the edge for basic RF signal identification. The participant will work with existing RF datasets, perform data preprocessing, and experiment with compact neural network models suitable for limited hardware. Tools may include Python, PyTorch, and simple embedded AI frameworks.
The work will focus on training and evaluating selected models rather than developing full production systems. The intern will compare accuracy and efficiency across approaches and document challenges that arise when applying deep learning to RF signals.
The expected outcome is a preliminary prototype model and a short evaluation summary describing its performance and potential next steps. This project provides hands on experience with applying AI methods to RF data while keeping the scope manageable within a 10 week timeframe.
Candidate skills that would help in this role include:
- Background in machine learning or deep learning, and proficiency in Python and common ML frameworks (such as PyTorch or TensorFlow)
- Experience with RF signal processing, wireless communications, or electromagnetic spectrum analysis
- Familiarity with model optimization techniques (quantization, pruning, knowledge distillation, etc.)
- Experience with edge or embedded computing platforms and understanding SWaP constraints
- Experimental, evaluation, and debugging skills for complex AI systems
Technical Skills
- Artificial Intelligence/Machine Learning
- Communication and Networking Technologies
- Computer Science
- Electrical Engineering
