Graduation Year

2026

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Electrical Engineering

Major Professor

Zhuo Lu, Ph.D.

Committee Member

Mia Naeini, Ph.D.

Committee Member

Nasir Ghani, Ph.D.

Committee Member

Guangjing Wang, Ph.D.

Committee Member

Xiaoyu Liu, Ph.D.

Keywords

Artificial Intelligence, Cybersecurity, Federated Learning, Mobile Computing, Radio Frequency Identification

Abstract

The rapid advancement of wireless networks has enabled a wide range of applications such as federated learning (FL) and Internet of Things, with supporting technologies like radio frequency identification (RFID). However, resource management and security challenges in wireless networks remain insufficiently addressed for these representative applications. In this dissertation, we first investigate wireless resource management for FL, then study novel attack and defense mechanisms for RFID systems. Finally, we analyze the performance and security of artificial intelligence (AI)-enabled next generation wireless networks.

First, we investigate the client assignment and bandwidth allocation problem for FL in a multi-provider wireless setting. Unlike existing models that assume a single provider, we consider a more practical and challenging scenario in which each client must be assigned to an appropriate wireless provider and allocated sufficient bandwidth. We formulate this problem as a non-linear, non-convex, combinatorial optimization problem with a cost constraint that limits bandwidth usage. The problem is addressed using our proposed algorithm and evaluated through comprehensive experiments.

Second, we investigate a novel attack and defense problem in RFID systems. Existing attacks on RFID mainly focus on confidentiality and accessibility, while integrity attacks have received much less attention. To address this gap, we propose a new integrity attack, called the identification overriding (IDO) attack, which exploits vulnerabilities in the widely adopted Electronic Product Code Generation-2 protocol for RFID. By manipulating the decoding process of the backscattered signal from the RFID tag, the proposed IDO attack does not require knowledge of the tag's transmitted data and reduces the injection signal power to make the malicious signal stealthier. Extensive experiments validate the effectiveness of IDO attack. In addition, we design a novel detection mechanism to identify the presence of such an attack.

Finally, we systematically study how AI can enhance both the performance and security of wireless networks. From the physical layer and media access control (MAC) layer to the network layer and higher layers, we examine how AI techniques can be applied to optimize network performance at each level. We also investigate the dual role of AI in enabling both attack and defense mechanisms in wireless networks. In addition, we discuss future challenges and potential research directions for AI-enabled wireless networks.

Download

Share

COinS