Practical security scheme design for resource-constrained wireless networks

The implementation of ubiquitous computing (or pervasive computing) can leverage various types of resource-constrained wireless networks such as wireless sensor networks and wireless personal area networks. These resource-constrained wireless networks are vulnerable to many malicious attacks that often cause leakage, alteration and destruction of critical information due to the insecurity of wireless communication and the tampers of devices. Meanwhile, the constraints of resources, the lack of centralized management, and the demands of mobility of these networks often make traditional security mechanisms inefficient or infeasible. So, the resource-constrained wireless networks pose new challenges for information assurance and call for practical, efficient and effective solutions. In this research, we focus on wireless sensor networks and aim at enhancing confidentiality, authenticity, availability and integrity, for wireless sensor networks. Particularly, we identify three important problems as our research targets: (1) key management for wireless sensor networks (for confidentiality), (2) filtering false data injection and DoS attacks in wireless sensor networks (for authenticity and availability), and (3) secure network coding (for integrity). We investigate a diversity of malicious attacks against wireless sensor networks and design a number of practical schemes for establishing pairwise keys between sensor nodes, filtering false data injection and DoS attacks, and securing network coding against pollution attacks for wireless sensor networks. Our contributions from this research are fourfold: (1) We give a taxonomy of malicious attacks for wireless sensor networks. (2) We design a group-based key management scheme using deployment knowledge for wireless sensor networks to establish pair-wise keys between sensor nodes. (3) We propose an en-route scheme for filtering false data injection and DoS attacks in wireless sensor networks. (4) We present two efficient schemes for securing normal and XOR network coding against pollution attacks. Simulation and experimental results show that our solutions outperform existing ones and are suitable for resource-constrained wireless sensor networks in terms of computation overhead, communication cost, memory requirement, and so on