← Back to Projects

Modulation Classification with Deep Learning

Train a neural network to listen to a radio signal and tell you which modulation scheme it uses.

Role
Course Project
Period
Sep. 2025 – Jan. 2026
Institution
SUSTech Supervised by Dr. Guang Wu

Why this matters

A receiver has to know the modulation scheme before it can demodulate. Hand-designed discriminative features work until the channel conditions change, then they break. Letting a network learn directly from the raw waveform is the more general approach.

What I did

  • Generated QPSK, 8-PSK, and 64-QAM baseband signals in MATLAB, including symbol mapping, upsampling, and square-root raised cosine (RRC) pulse shaping.
  • Analyzed time-domain and frequency-domain characteristics of modulated signals using FFT to study bandwidth and spectral properties.
  • Applied a convolutional neural network (CNN) to classify digital modulation schemes and interpreted classification probabilities.

Technologies

  • MATLAB
  • Digital Communications
  • FFT Spectral Analysis
  • Convolutional Neural Networks