|
Title
|
|
|
|
Resource allocation for intelligent reflecting surface aided wireless networks
| |
|
Author
|
|
|
|
| |
|
Abstract
|
|
|
|
| In today's world, staying connected is more important than ever, but achieving reliable wireless communication everywhere can be a challenge. This dissertation introduces a cutting-edge technology known as Intelligent Reflecting Surfaces (IRSs) that promises to revolutionize how we connect. Imagine a smart, invisible “mirror” that can bend and direct wireless signals precisely where needed, overcoming obstacles and ensuring your device always gets a strong connection. That is what the IRS does. IRS, at its core, is a sophisticated planar array, composed of numerous passive or active elements capable of individually manipulating electromagnetic waves to reshape the wireless signal propagation environment. By smartly adjusting the phase and amplitude of these elements, an IRS can seamlessly steer signals toward intended receivers, effectively creating optimized communication paths even in scenarios where direct Line-of-Sight (LoS) is obstructed. This ability to mold the propagation environment on demand, without additional energy for signal transmission, enables the IRS to enhance connectivity in diverse environments, from densely built urban areas to indoor spaces. Furthermore, the ability of the IRS to operate without the need for active power amplification allows for a significant reduction in energy consumption, making it an eco-friendly solution for extending and improving wireless network coverage. In this dissertation, IRS is presented as a key enabler for a myriad of advanced technologies, unlocking new potentials across various high-tech fields by enhancing their performance and efficiency. By strategically manipulating electromagnetic waves, IRS provides a solution to enhance power efficiency in multi-user Simultaneous Wireless Information and Power Transfer (SWIPT) networks. This capability allows for a steady flow of information and power transfer, illustrating the dual capability of the IRS to support energy harvesting and data transmission. Furthermore, the integration of IRS into Ultra-Reliable Low-Latency Communication (URLC) and Machine Type Communication (MTC) systems emerges as a game-changer, significantly reducing latency and increasing reliability. IRS can significantly benefit Virtual Reality (VR) users facing considerable path loss or blockages, ensuring immersive experiences without latency or loss of quality. IRS also enhances Mobile Edge Computing (MEC) by optimizing signal delivery for efficient edge data processing. These improvements are essential for critical applications requiring instantaneous feedback and high levels of data integrity, such as autonomous vehicles and industrial automation, underpinning the role of the IRS in facilitating the next wave of communication needs. This work delves into the strategic deployment of IRS across a broad frequency spectrum, from Frequency Range 1 (FR1) to Frequency Range 2 (FR2), extending into the higher frequency domains of millimeter-Wave (mmWave) and TeraHertz (THz) frequencies, illustrating its profound impact on the future of telecommunications. In order to investigate the performance of IRS-assisted networks, this dissertation defines a range of Key Performance Indicators (KPIs), such as data rate, power efficiency, energy efficiency, Signal-to-Interference-plus-Noise Ratio (SINR), transmit signal power budget, and received power strength. These KPIs serve as metrics to assess and optimize the network's performance based on designing an efficient resource allocation policy. Non-linear, nonconvex, and Mixed Integer Nonlinear Programming (MINLP) problems arise when addressing the resource allocation optimization problem. These problems are Non-deterministic Polynomial time (NP)-hard due to the complex relationship between variables and the system's constraints. Given the complexity of these optimization problems, different strategies are used to simplify and approach their solution. By relaxing the objective function (the NPs) and constraints that are non-convex to a more tractable format, the problems became more manageable. This relaxation approach often involved transforming the optimization problem into its convex equivalent or utilizing approximation techniques to linearize or convexify non-convex terms. Algorithms are developed that are capable of solving the main problem either globally or suboptimally but sufficiently close to the global optimum. These solutions employ optimization solvers and computer simulations, exploiting advanced mathematical tools and techniques such as the big-M method for linearizing product terms involving binary variables and Successive Convex Approximation (SCA) to obtain convex approximations of non-convex terms. The iterative nature of these solutions allowed for step-by-step refinement, gradually moving towards an optimal configuration of a resource allocation design despite the initial problem's complexity. Through exhaustive simulations, this dissertation unveils the diverse performance improvements achievable through resource allocation in IRS-assisted networks, providing rich insights into how IRS technology can improve wireless communication systems. These simulations serve as a critical bridge, connecting theoretical predictions with empirical evidence and validating the practical feasibility of the proposed IRS-enhanced network. By exploring various IRS configurations — examining both passive and active types and varying the number of reflective elements — and their implementation in different environments and settings, this study not only confirms the theoretical models' accuracy but also explains the conditions under which IRS deployments yield maximal performance gains, manifesting the IRS versatility in adapting new technologies. Collectively, this dissertation studies the impact of IRS across a broad range of technologies. By enhancing the performance of SWIPT networks, facilitating URLLC and MTC, enabling MEC, and revolutionizing VR, mmWave, and THz applications, IRS stands at the forefront of wireless communication innovation. This work demonstrates the diverse applications of IRS technology and lays the foundation for future research aimed at utilizing IRS to tackle the dynamic challenges of modern wireless networks. It charts a path toward the development or more robust, efficient, and engaging communication ecosystems. |
| |
|
Language
|
|
|
|
English
| |
|
Publication
|
|
|
|
Antwerp
:
University of Antwerp, Faculty of Applied Engineering
,
2024
| |
|
ISBN
|
|
|
|
978-90-5728-855-5
| |
|
DOI
|
|
|
|
10.63028/10067/2069080151162165141
| |
|
Volume/pages
|
|
|
|
xxvi, 284 p.
| |
|
Note
|
|
|
|
:
Famaey, Jeroen [Supervisor]
:
Berkvens, Rafael [Supervisor]
| |
|
Full text (open access)
|
|
|
|
| |
|