Intrusion Tolerant Routing Protocols for Wireless Sensor Networks

Abstract

This MSc thesis is focused in the study, solution proposal and experimental evaluation of security solutions for Wireless Sensor Networks (WSNs). The objectives are centered on intrusion tolerant routing services, adapted for the characteristics and requirements of WSN nodes and operation behavior. The main contribution addresses the establishment of pro-active intrusion tolerance properties at the network level, as security mechanisms for the proposal of a reliable and secure routing protocol. Those properties and mechanisms will augment a secure communication base layer supported by light-weigh cryptography methods, to improve the global network resilience capabilities against possible intrusion-attacks on the WSN nodes. Adapting to WSN characteristics, the design of the intended security services also pushes complexity away from resource-poor sensor nodes towards resource-rich and trustable base stations. The devised solution will construct, securely and efficiently, a secure tree-structured routing service for data-dissemination in large scale deployed WSNs. The purpose is to tolerate the damage caused by adversaries modeled according with the Dolev-Yao threat model and ISO X.800 attack typology and framework, or intruders that can compromise maliciously the deployed sensor nodes, injecting, modifying, or blocking packets, jeopardizing the correct behavior of internal network routing processing and topology management. The proposed enhanced mechanisms, as well as the design and implementation of a new intrusiontolerant routing protocol for a large scale WSN are evaluated by simulation. For this purpose, the evaluation is based on a rich simulation environment, modeling networks from hundreds to tens of thousands of wireless sensors, analyzing different dimensions: connectivity conditions, degree-distribution patterns, latency and average short-paths, clustering, reliability metrics and energy cost.