Abstract
This thesis addresses fundamental security challenges in heterogeneous IoT environments, where devices using different wireless technologies (WiFi, ZigBee, and Bluetooth) must interact securely while maintaining network interoperability. As these heterogeneous networks become increasingly interconnected, establishing secure and efficient authentication mechanisms across different technologies has become crucial.First, we investigate security vulnerabilities in Cross-Technology Communication (CTC), which serves as a foundational bridge between heterogeneous networks. We propose a lightweight detection model that enables commodity ZigBee devices to identify unauthorized CTC signals by analyzing inherent chip error patterns. This approach establishes a first line of defense in heterogeneous environments without requiring specialized hardware, incorporating data augmentation and model pruning techniques to ensure robust performance across diverse network conditions.
Building on this security foundation, we present AUTHFi, a comprehensive authentication framework designed specifically for heterogeneous IoT environments. AUTHFi enables commodity WiFi devices to authenticate IoT devices using different wireless technologies, eliminating the need for expensive Software-Defined Radios (SDRs). The framework introduces innovative techniques for signal compensation and Carrier Frequency Offset (CFO) estimation, along with a fusion neural network that effectively combines physical-layer features across different technologies for reliable authentication.
Finally, we develop a blockchain-based secure, interactive, and fair Mobile Crowdsensing (MCS) framework that addresses the broader challenges of secure interaction in heterogeneous networks. This system implements cross-technology identity verification using private keys and location-based symmetric key generation, while efficiently managing authentication across different network domains. Our framework incorporates a Stackelberg game model to ensure fair resource allocation and rewards across heterogeneous network participants.
The innovative approaches presented in this thesis make significant theoretical and practical contributions to securing heterogeneous IoT environments, providing a comprehensive foundation for developing more secure, efficient, and scalable cross-technology authentication systems.
| Date of Award | 24 Apr 2025 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Zhimeng YIN (Supervisor) |