A Knowledge Base Parameter (KBP) Based Spray and Wait Routing Protocol for Opportunistic Network

Abstract

Intermittently connected networks are a special type of mobile ad-hoc networks where the connectivity from source node to destination node is unlikely exist. Transmission of a message in such networks is happens by store-carry-forward strategy. In this way, message delivery depends on mobility of the nodes, contact opportunities with other nodes and the contact patterns. In opportunistic network there are several protocols exists which are broadly classified in the category such as forwarding mechanism based, replication based, control replication based and knowledge based. One of the control replication based protocol is binary spray and wait. The binary spray and wait gives good performance compare to others in many scenarios but still it require some improvements such as selection of initial number of message copies, information discrimination technique and selection of important node to maximize the delivery ratio with minimum numbers of message drop and less overhead ratio. Proposed work is to implement an efficient variant of spray and wait protocol for information discrimination in opportunistic network. Instead of blind spraying, this variant takes minded decision using the knowledge based mechanism to identify the good candidate for spraying messages in the network. In proposed protocol every node is required to maintain the past rate of encounter history, which is used to predict future encounter rate and forwards message copies according to encounter value. Using the Opportunistic Network Environment (ONE) simulator, experimental work has been carried out under real and synthetic dataset which shows that variant is performing better compare to binary spray and wait protocol as well as other conventional protocols.