Gaussian Processes for Nonlinear Signal Processing
This is an incremental tutorial for signal processing practitioners, aiming to bridge a gap between machine learning and signal processing methods.
The paper tackles the underuse of Gaussian processes in signal processing by presenting them as a nonlinear extension to Wiener filtering, covering aspects like adaptive algorithms and applications in wireless communications.
Gaussian processes (GPs) are versatile tools that have been successfully employed to solve nonlinear estimation problems in machine learning, but that are rarely used in signal processing. In this tutorial, we present GPs for regression as a natural nonlinear extension to optimal Wiener filtering. After establishing their basic formulation, we discuss several important aspects and extensions, including recursive and adaptive algorithms for dealing with non-stationarity, low-complexity solutions, non-Gaussian noise models and classification scenarios. Furthermore, we provide a selection of relevant applications to wireless digital communications.