Secure and reliable resource allocation in multi-function wireless systems

• Main Author: Lotfi, Ismail
• Other Authors: Dusit Niyato
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/171744
• Institution: Nanyang Technological University

Multi-function wireless systems offer numerous benefits, such as efficient spectrum re-utilization and minimized hardware costs, by enabling multiple tasks to be performed simultaneously using the same spectrum or device. This has led to the widespread adoption of multi-function devices by various entities. However, these devices can be vulnerable to external threats, such as jammers or wardens, and can also be a source of attacks against the service provider with which they interact. Therefore, it is crucial to address the security and reliability challenges of multi-function wireless systems from different angles and perspectives. This thesis focuses on studying the security and reliability threats in multi-function wireless systems at three layers: the physical layer, networking layer, and application layer. The first part of this thesis focuses on the reliability issue of multi-function wireless systems at the physical layer. We present a novel system design that mitigates jamming attacks using deep reinforcement learning (DRL). Our design not only resists jamming attacks but also improves system performance by intelligently leveraging jamming signals. We employ backscatter technology and deception strategy to use jamming attacks on multi-function wireless systems. Backscatter technology transmits data on jamming signals, while the deception strategy predicts the jammer’s action and adopts the appropriate counterattack instantaneously. Our DRL-based system design demonstrates that our proposed multi-function wireless system design is secure and reliable against different types of jamming attacks. The second part of this thesis focuses on the security aspects of multi-function wireless systems at the physical and networking layers. At the physical layer, we propose a covert multi-function wireless system for joint radar and communication (JRC) applications. In the networking layer, we design a robust multi-item multi-buyer auction mechanism for channel allocation that protects the mobile operator from any misbehavior by the multi-function nodes. This auction mechanism addresses the uncertainty of the warden’s location while friendly jammers are deployed to maximize the covertness of the transmitted signals. The robustness of this multi-item multi-buyer auction system ensures secure and reliable channel allocation in multi-function wireless systems. The third part of this thesis focuses on a system-level application that aims to protect a virtual service provider from attacks by malicious multi-function nodes in the wireless system. We propose a learning-based iterative contract based on multiagent reinforcement learning that helps the service provider incentivize wireless nodes to participate truthfully in the contract bundle derivation process. Our framework effectively mitigates the adverse selection problem in contract theory with minimal requirements for disclosing the private types of the participants. We demonstrate that the proposed framework has interesting applications beyond multi-function wireless systems and contract theory. In summary, this thesis addresses various security and reliability challenges in emerging multi-function wireless systems from multiple perspectives. We develop novel system designs and mechanisms, validated through extensive simulations, that provide valuable insights and findings. Our work enables promising applications of multi-function wireless systems, paving the way for a more secure and reliable wireless future.