ÎÞÓÇ´«Ã½

This course provides a survey-level introduction to wireless communications, RF signals, RF signals collection, and technical analysis of RF signals. We will observe the basic principles of RF signals through experimentation with software-defined radios (SDR).​

This course provides the foundation in math and theory of signals analysis, including analog-to-digital conversion and sampling/aliasing, Fourier analysis and transforms, filter design, noise properties and reduction, probability and statistics, and linear algebra. Participants should have a good theoretical foundation in RF signals analysis, generating visualizations, simulating basic transmitter/receiver models, and evaluating signal/noise metrics.

This course introduces detection theory for RF signals. After this course, participants given an RF recording will be able to apply different detection techniques to detect signals present.

This course introduces pulsed signals and analysis techniques. We will generate proof of concept demonstration using SDR to transmit and identify different pulsed signals.

This course provides a comprehensive understanding of machine learning (ML) for RF signals detection, matched filter and pulsed signals, probability theory for signals detection, cyclostationary processing for signals detection, and overall detection improvement methods.

• Create linear frequency modulated (LFM), binary phase-shift keying (BPSK), and noise signals of varying signal-to-noise ratios (SNR) – create a labeled machine learning dataset of these signals 
• Perform signal analysis on sample signal files for each class: time, frequency, constellation, spectrogram plots, calculate SNR
• Create spectrogram images of each signals file
• Leveraging TensorFlow/Keras, train and test a spectrogram-image-based signal classifier neural network
• Format code for this project into a well-documented project in Github, with Python package requirements documented, and a diagram of the data flow for your ML pipeline  ​