Dynamics of social-aware pervasive networks

Social-aware pervasive networks consider the users' social behavior to overcome intermittent end-to-end connectivity, inherent to this type of networking: forwarding decisions consider local knowledge about the behavior of nodes to predict future encounters. Complex Network Analysis (CNA) has been used to support contact prediction, by aggregating connectivity graphs into less volatile social graphs. Nevertheless, the structure of such graphs is rather dynamic, since users' social behavior and interactions vary throughout their daily routines and according to their mobility. Consequently, aggregation algorithms should be able to create social graphs that reflect the resulting dynamic behavior of people. This paper discusses on human behavior-aware aggregation to allow the creation of graphs based on social variations observed in people's daily routines. By focusing on the dynamics of the network, we show that social graphs, reflecting different stages of human social behavior and mobility, are able to take advantage of the potential small-world properties of networks in different time frames, improving the performance of social-aware opportunistic forwarding.