Adaptive Spray: An Efficient Restricted Epidemic Routing Scheme For Delay Tolerant Networks

In Delay Tolerant Networks, there is no guarantee that a fully connected path between source and destination exists any time. In this context, the limited hop count scheme which allows a message to travel at most k hops is considered as a basic routing scheme. However, there are no significant works offering an efficient method to identify a proper value for k. In this paper, we propose a new approach, named Adaptive Spray (AS-scheme). Instead of identifying a strict value k, it heuristically identifies a real-value coefficient to improve the network performance while eliminating the computational complexity. By leveraging only local knowledge like messages remaining time and nodes meeting rate, AS-scheme is highly scalable. Simulation results reveal that the proposed approach can quickly identify an appropriate coefficient and outperforms Epidemic algorithm in terms of increasing delivery rate, reducing overhead ratio and average latency, especially in dense networks.