Delay tolerance for constrained IPv6 networks

Low power sensor networks have traditionally been regarded as not having the capabilities required to connect them to the internet. New research into the Internet of Things has challenged this concept and is opening up new possibilities for sensor network capabilities. Environmental sensor networks are just one of the areas which will greatly benefit from this connectivity improvement. However, there are many challenges to be solved in order to make full and efficient use of these advancements.One of the major challenges which has been identified is the lack of connectivity when sensors are in low power sleep states. Previous solutions for low power devices have relied on application layer gateways to proxy communications to the sensors, but this restricts the flexibility of the network as it is limited to the capabilities of the proxy. Delay Tolerant Networking (DTN) offers a solution to this problem by allowing sensors to respond and handle communications at their convenience.This thesis presents and evaluates a novel method and implementation of Delay Tolerant Networking using IPv6 extension headers. The proposed DTN extension header is found to have a significantly lower packet size overhead than other DTN protocols. In addition, the protocol and systems to support it are entirely backwards and forwards compatible with the existing internet infrastructure allowing for it to be incorporated into existing deployments. The developed protocol forms a new state of the art for DTN on constrained sensor networks using end to end IP connectivity. Using this, a new range of low power IoT devices can be developed, featuring long battery lives and reliable connectivity