Friendly-drop: A social-based buffer management algorithm for opportunistic networks

In the last five years, the integration of social relationships in algorithms for Opportunistic Networks (OppNet) has received much attention from researchers worldwide. Despite node mobility has a great variability along time and is often unpredictable, node relationships can be less volatile than node mobility. In this work, we investigate the impact of using social relationships for managing the buffer of OppNet nodes. First, we classify the relationships between the nodes in two groups: their friendship, and their acquaintanceship. Then, we propose the Friendly Drop Algorithm (FDA), which combines in its decisions both the friendship among the nodes and their acquaintanceship. FDA explores the self-reported friendship relationships from the nodes to build friendship graphs. Moreover, FDA uses a metric based on the contact similarity among the nodes to find out their acquaintanceship. We validated the proposed algorithm by using trace-driven simulations through the Opportunistic Network Environment simulator (The ONE). Experiments show that FDA contributes to increase the delivery ratio and decrease the number of forwardings associated to standard replication algorithms (e.g. Epidemic). We also find that combining both forwarding algorithms and buffer management algorithms based on social characteristics can improve the network performance.