Statistical Signal Processing and Supervised Learning (PhD)

The first part of the course reviews the fundamentals of statistical signal
processing with insight on deep learning methods, the second part is
focused to selected areas where the theory methods are paired to some
applications chosen to be general enough to provide a useful background
for many interdisciplinary contexts such as audio and digital
communications, imaging and machine learning, navigation and
estimation in networks (26h)

  • Fundamentals of estimation theory (BLUE, MLE, CRB, MMSE, MAP)
  • Parameter tracking and positioning, Kalman filtering (KF, EKF, UKF)
  • Classification of signals and supervised deep learning methods

Applications:

  • Prediction of portfolio balance and E-Marketing prediction
  • Inertia parameters (accelerometers and gyroscopes) tracking and high-
    resolution positioning
  •  Array signal processing and multichannel MIMO systems