Towards social-aware opportunistic network datasets

Abstract

Delay-tolerant networks are wireless networks designed to be used in cases where network infrastructure is nonexistent or not available to be used. Because of this, there are several problems that need to be addressed in this environment, such as lack of continuous end-toend connectivity and increased delay and error rates in data transfer. As such, conventional routing schemes aren't feasible in providing e cient solutions for these cases. Since the nodes present in these kinds of networks usually possess very limited resources, opportunistic routing protocols should not only try to achieve a good message delivery probability, but also reduce the number of message replicas present in the network. This is done so as to avoid an unnecessary waste of storage and energy that comes from storing and transmitting messages to other nodes. Some of the recent Delay-tolerant network routing proposals involve using social information to determine which node has a higher probability of successfully delivering a message to its intended destination. This seems to be a popular strategy, that achieves a good delivery probability while reducing the message overhead, when compared to simpler schemes. One way to analyze the performance of a routing protocol is to use real opportunistic contact datasets to simulate a real life environment. This work focuses on providing a research on opportunistic network traces as a way to determine the contact patterns of Delay-Tolerant network nodes and their impact on routing algorithm performance, as well as proposing an architecture for a future data collection experiment.