Opportunistic Network Behavior in a Swarm: Passing Messages to Destination

Published: 2020
Proceedings of the 5th Asia-Pacific Conference on Intelligent Robot Systems (ACIRS 2020)
ISBN: 978-1-7281-9818-7

Abstract

This paper evaluates and compares two routing algorithms for message passing in opportunistic peer-to-peer networks in the swarm network simulator Swarm-Sim. The two considered algorithms, Epidemic Routing and PRoPHET Routing, are compared in terms of delivery rates, efficiency and overhead, while presenting different mobility models. By this, we combine a variation of mobility models and a set of communication protocols, measuring their impact on the swarm behavior. Message delivery rates for Epidemic routing are considerably and consistently better. However, PRoPHET generates less overhead than Epidemic Routing independent of mobility models and scenarios. The measured overhead of Epidemic exceeded that of PRoPHET by about 100%. Memory requirements for keeping delivery probabilities - used by PRoPHET to circumvent flooding - might be a limitation as they must be kept locally per peer, this detail offers aspects for further research in the future. Finally, this paper demonstrates Swarm-Sim to be a viable alternative to simulate routing protocols in opportunistic peer to peer networks without requiring great insight into the simulator’s architecture and thus establishing itself as an easy to use simulator. It achieves simplicity without yielding any of flexibility of application.