DTN Routing Algorithm in Temporary Shelter based on Mobility Social Attributes and Message Destination Prediction

Opportunistic Mobile Social Networks (OMSNs) represent a unique class of Delay-Tolerant Networks (DTNs) comprised of mobile nodes equipped with communication devices. Data transmission within these networks hinges on interactions between mobile nodes, characterized by unstable connectivity and frequent node mobility. Existing DTN routing algorithms, which incorporate social attributes such as degree centrality, contact frequency, and connection duration, build upon traditional approaches. However, many of these algorithms overlook the mobile social characteristics of nodes, limiting their practical applicability in real-world scenarios. In this work, we utilize the SMOOTH and SPMBM mobility models to simulate realistic movement trajectories in a temporary shelter context, while deeply exploring the mobile social attributes of nodes. We propose a methodology for categorizing friend relationships based on node contact scenarios and predict the mobility of message destinations using node friendships. Our validation results indicate a prediction accuracy of 70% in a GPS-denied environment. Furthermore, through rigorous controlled experiments, our approach has demonstrated superiority over comparable algorithms in terms of delivery rate, network load, and forwarding hop counts.