Optimising DTN routing protocols for efficient use in disaster scenarios

Disasters occur anywhere around the world at unexpected times and they cannot usually be predicted or avoided. When a disaster arises, a significant proportion of the surrounding population is devastated. One of the most important features of coping with any disaster is maintaining communication, as it is essential for both management and recovery. Traditional forms of communication may not be possible due to the nature of the disaster; it is likely that the previous network infrastructure will be destroyed. Delay Tolerant Networks (DTNs) can provide a feasible solution. DTNs consist of mobile nodes that can communicate with each other even if a route connecting them never exists. They do not need an end-to-end connection and routes can be dynamically built when and where they are required. This makes them applicable for deployment in disaster scenarios. DTNs use protocols to deliver messages to their destinations, however the existing protocols have numerous problems. The existing DTN routing protocols do not prioritise messages based on their content. This concept leads to the idea of optimising an existing protocol to forward messages based on their urgency as opposed to their existing structure. Various disaster simulations were created using the Opportunistic Network Environment (ONE) Simulator, which can be used to simulate and evaluate how DTN routing protocols disseminate their messages. ONE was used to test the overall performances in order to see if the optimised protocol performed better than the existing version in the given scenarios. The resulting protocol (Disaster Optimised Spray and Wait) significantly increased the delivery probability of high urgency messages while maintaining the overall delivery probability. The average latency was slightly increased, yet this increase was essential and justified. The overhead ratio remained relatively constant and did not hinder the overall performance. Therefore an efficient protocol was produced, which would in theory be suitable for deployment in countless numbers of disaster scenarios