Secure Routing in Multi-hop Wireless Networks

Routing is a fundamental networking function in every communication system, and multihop wireless networks are no exceptions. Attacking the routing service, an adversary can easily paralyse the operation of an entire network. Compared to traditional wired networks, such attacks can be performed relatively easily in wireless networks due to the unsupervised access to the wireless medium. The malicious manipulation of some routing messages results in the dissemination of incorrect routing information which can eventually lead to network malfunction. Even more, intermediate nodes can be corrupted, and thus, exhibit arbitrary behaviour. Considering these facts, securing routing protocols is a primary task, however, designing such secure routing protocols is not a straightforward procedure. A widely used method has been so far to identify different types of possible attacks against routing, and to define routing security implicitly as resistance to these attacks. However, this approach does not provide a common ground for comparing routing protocols in terms of security. Moreover, due to the subtle nature of attacks against routing protocols, such informal reasoning is an error-prone method. In this dissertation, I develop a formal framework in which precise definitions of secure routing can be given, and secure routing protocols proposed for multi-hop wireless networks can be rigorously analysed. I demonstrate the usefulness of this framework in several ways: first, I prove the security of several existing routing protocols that were proposed earlier independently from my work. I show that my model is capable of distinguishing between routing protocols in terms of security. Second, applying the design principles that were identified during the analyses, I propose novel routing protocols for wireless ad hoc and sensor networks and I prove that they are secure in my model.