Abstract
Delay/Disruption Tolerant Networks (DTNs), also known as Intermittently Connected Networks (ICNs), are wireless networks in which there is no contemporaneous end-to-end connectivity. In these networks, the message delivery relies on the Store-Carry-Forward (SCF) routing behavior. Routing is an important research area in DTNs. Up to now, the partial knowledge about network topology information has been adopted by numerous previous works using encounter prediction. Since the large variation of network topology is a key challenge for routing in DTNs due to sparse network density, the motivation for proposing geographic routing in DTNs is because of the least dependence on network topology information. In addition, geographic routing inherently does not rely on the contemporaneous end-to-end connectivity to relay the message. Considering that geographic routing in DTNs has not received much attention, the contributions in this thesis are as follows: 1:- The motivation for investigating geographic routing in DTNs together with its challenges are identified, based on an original taxonomy on routing in DTNs. The challenges are considered as how to reliably relay the message, handle the local maximum problem, and address the unavailable realtime location of mobile destination. 2:- A geographic routing algorithm considering the stationary destination is proposed, focusing to overcome the limitation of the given geometric metric requiring both pairwise nodes moving towards destination and handle its local maximum problem. 3:- Based on overcoming these two challenges, a geographic routing algorithm using historical geographic information of the destination is proposed, by taking into account its mobility. This overcomes the challenge to obtain the current geographic information of mobile destination due to sparse network density, as the centralized location service system may be unavailable in this case. 4:- Finally, a delegation based geographic routing algorithm is proposed, while overcoming the identified three challenges. The motivation is to further enhance the routing performance in an efficient way.