A Fuzzy Ranking Based Buffer Replacement Strategy for Opportunistic Networks

Abstract

In opportunistic networks, it always suffers from the problems that the topology of the network changes dramatically, the disconnection between nodes are prevalen. A store-carry-forward protocol is utilized to deal with those problems in which messages need to be stored in buffers and carried for a long time. Allowing for the limited buffer space, it likely brings buffer overflows and network congestion. A buffer replacement strategy based on fuzzy ranking is proposed to ease congestion and improve efficiency for opportunistic networks in this paper. In the strategy priority, timeliness and size of messages are taken into account, and fuzzy theory is employed to make multi-attribute evaluation and rank replacement sequence. Through simulation results, it can be observed that the proposed strategy can improve the effect of routing in opportunistic networks in terms of delivery ratio and packet loss rate.