Gaussian Process Techniques for Wireless Communications

TL;DR

This paper explores Gaussian process techniques within wireless communications, focusing on Bayesian filtering methods like Kalman and Particle filters, and proposes a new probabilistic inference methodology for non-linear models.

Contribution

It introduces a novel, general Gaussian process-based approach for inference and learning in non-linear state-space models, enhancing existing Bayesian filtering methods.

Findings

Classical filters like Kalman and Particle filters are discussed.

Gaussian processes are used as a non-parametric estimation technique.

A new methodology for inference in non-linear models is proposed.

Abstract

Bayesian filtering is a general framework for recursively estimating the state of a dynamical system. Classical solutions such that Kalman filter and Particle filter are introduced in this report. Gaussian processes have been introduced as a non-parametric technique for system estimation from supervision learning. For the thesis project, we intend to propose a new, general methodology for inference and learning in non-linear state-space models probabilistically incorporating with the Gaussian process model estimation.