Date of Award

5-2026

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Electrical Engineering and Computer Science

First Advisor

Neji Mensi

Second Advisor

Md Selim Habib

Third Advisor

Sujoy Ghosh Hajra

Fourth Advisor

Brian A. Lail

Abstract

The digital era is being driven by an expanding ecosystem of transformative technologies, such as the Internet of Things (IoT), artificial intelligence (AI), virtual reality (VR), augmented reality (AR), and the metaverse. These applications collectively demand wireless networks that can support billions of simultaneously connected devices with ultra-high data rates, low latency, and reliable coverage. However, conventional wireless systems are fundamentally insufficient to meet these requirements. Consequently, non-orthogonal multiple access (NOMA) and simultaneous transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) have been proposed. NOMA allows multiple users to share the same resource block (time/frequency/code) in the power domain. In addition, STAR-RIS creates a programmable and controllable propagation environment where high-frequency signals can overcome the high blockages. The integration of STAR-RIS with NOMA creates a powerful architecture that achieves large-scale IoT connectivity and improved quality-of-service (QoS). As a result, this integration stands as one of the most promising solutions for next-generation wireless networks. Since the STAR-RIS can serve the users within 360◦ full coverage, the potential eavesdroppers can intercept the confidential information across both the transmission and reflection sides. Therefore, to mitigate the impact of eavesdropping attacks, physical layer security (PLS) has been recognized as one of the most effective solutions for wireless systems. PLS ensures confidentiality by exploiting wireless propagation characteristics such as propagation mediums, fading, and channel properties. In contrast to advanced cryptographic techniques, PLS is particularly suitable for lightweight and low-power IoT devices, where encryption is impractical to implement. In this context, this thesis investigates the secrecy outage probability (SOP) and intercept probability (IP) in order to analyze the secrecy rate guarantees and the interception risks over Nakagami-m fading channels in a STAR-RIS-assisted NOMA wireless system. Building on these findings and recognizing that the mathematical analysis provides the theoretical foundation for further optimization, this thesis further investigates the secure transmission in a STAR-RIS-assisted downlink multiple-input single-output (MISO) wireless network. The secrecy rate is maximized by the joint design of the transmit beamforming and the transmission and reflection coefficients of the STARRIS, while satisfying the electromagnetic property of the STAR-RIS and the transmit power limit of the base station. Since this communication network is in a dynamic environment, the optimization problem is non-convex and mathematically difficult to solve. To address this issue, the deep reinforcement learning (DRL)-based algorithm, namely soft actor-critic with directed exploration (SAC-DE), is proposed to obtain the maximum reward by continuously interacting with and learning from the dynamic environment.

Comments

Copyright is held by author.

Download

Available for download on Tuesday, May 09, 2028

Share

COinS