ORCID
https://orcid.org/0000-0002-1917-2158
Keywords
Localization, mmWave Beam Prediction, Deep Learning
Abstract
The primary focus of this doctoral dissertation is the investigation of robust massive Multiple Input Multiple Output (MIMO) communication systems. The study centers on two closely related issues: i) single-site user localization, and ii) millimeter Wave (mmWave) beam prediction. The motivation is to develop a system that simultaneously offers users reliable communication and precise localization using a single infrastructure. This dissertation specifically addresses three key challenges in data-driven solution: i) the limited availability of position-labeled training datasets, ii) robustness in time-varying dynamic environments, and iii) the generalization of models to unseen scenarios. The study introduces three localization and two beam prediction frameworks each addressing at least one of the key challenges. For localization, the first framework extracts propagation path information from Channel State Information (CSI) samples and integrates it with the environment map to locate the user, eliminating the need for labeled datasets. The second framework reduces the requirement for labeled training data in conventional data-driven models by training a Gaussian Process Regression model on a small dataset of location-labeled CSIs to predict user locations. The third framework employs deep learning sequence prediction to detect and correct the CSI alterations caused by dynamic environmental changes. In the context of mmWave beam prediction, the first framework utilizes sub-6 GHz CSI samples to predict mmWave beams through a deep neural network in dual antenna systems, emphasizing robustness and generalization in dynamic environments. The second framework leverages multimodal deep learning to fuse sensor data collected from the environment to perform prediction. The framework combines blockage detection and beam prediction to enhance mmWave beam prediction performance in complex, dynamic settings.
Completion Date
2024
Semester
Fall
Committee Chair
Nazanin Rahnavard
Degree
Doctor of Philosophy (Ph.D.)
College
College of Engineering and Computer Science
Department
Electrical and Computer Engineering
Degree Program
Electrical Engineering
Format
Identifier
DP0029050
Language
English
Release Date
12-15-2024
Access Status
Dissertation
Campus Location
Orlando (Main) Campus
STARS Citation
Vuckovic, Katarina, "Robust Machine Learning Based Frameworks for Localization and mmWave Beam Prediction in Massive MIMO Systems" (2024). Graduate Thesis and Dissertation post-2024. 82.
https://stars.library.ucf.edu/etd2024/82
Accessibility Status
PDF accessibility verified using Adobe Acrobat Pro Accessibility Checker