Disseminação de mensagens em redes oportunistas baseada em relações sociais e aprendizagem de máquina

Abstract

Opportunistic networks provide communication support in areas with overloaded cellular networks, and in scenarios where a fixed infrastructure is not available, as in remote and developing regions. A critical issue, which still requires a satisfactory solution, is the design of an efficient data delivery solution that considers delivery efficiency, delay, and cost. To tackle this problem, most researchers have used either the network state or node mobility as a forwarding criterion. Solutions based on social behaviour have recently being considered as a promising alternative. Following the philosophy from this new category of protocols, in this work, we present the "FriendShip and Acquaintanceship Forwarding"(FSF), a routing algorithm that makes its routing decisions considering the social ties between the nodes and both the selfishness and the device resources levels of the candidate to message relay. When a contact opportunity arises, FSF first classifies the social ties between the message destination and the candidate to relay. Then, by using logistic functions, FSF assesses the relay node selfishness to consider those cases in which the relay node is socially selfish. To consider those cases in which the relay node does not accept to receive the message because its device has resource constraints at that moment, FSF looks at the resource levels of the relay node. Regarding the buffer management problems, in this work we introduced the algorithm Friendly-Drop (FDA) which takes its decisions based on the nodes social relationships. When the nodes buffer is with constraints, FDA drops the messages addressed to nodes with weak social relationship. On the other hand, if a pair of nodes are in contact, FDA sends the messages addressed to nodes having strong social relationships. By using the ONE simulator to carry out trace-driven simulation experiments, we have found that even considering the selfishness on routing issues, our FSF algorithm outperforms previously proposed schemes, by increasing the delivery ratio up to 20% with the advantage it needs a lower number of forwarding events. We have also found that the buffer management algorithm can become an important key to improve network performance in scenarios with selfish nodes.