LAUSR

Opportunistic network enables users to form an instant network for data sharing, which is a type of Ad-hoc network in nature, thus depends on cooperation between nodes to complete message transmission. Because of intermittent communication and frequent changes of topology structure in opportunistic networks, the duration of node encounters is limited, as well as the length of established connections. If the amount of interactive data is large and the communication bandwidth is small, the messages that need to be transmitted are not guaranteed to be delivered successfully every time. In this regard, this paper establishes a transmission prediction mechanism exploiting comprehensive node forwarding capability (TPMEC) in opportunistic networks. When quantifying the forwarding capability of nodes, the algorithm not only considers the cooperative tendency but also discusses the encounter strength between nodes. At the same time, in order to find out all key nodes during the transmission process, the algorithm adopts the theory of matrix decomposition to predict and supplement the missing forwarding capability value of nodes, thus improving the efficiency of message transmission. Simulation results show that compared with ITPCM algorithm, ETNS algorithm, Spray and Wait algorithm, and PRoPHET algorithm, the proposed scheme has the highest transmission success ratio and the lowest routing overhead.