Topology-aware Path Planning for In-Transit Coverage of Aerial Post-Disaster Communication Assistance Systems

Abstract

Over the last years, natural disasters have shown to impair and destroy communication infrastructure. This results in an increased importance of infrastructure-independent ad hoc communication systems, such as delay-tolerant networks (DTNs). Research has shown that these networks are able to provide basic communication functionality for civilians. However, they are limited in their performance as the network topology is highly intermittent due to the human nature of clustering around important locations like shelters and moving in groups. Small Unmanned Aerial Vehicles (UAVs) have proven to be efficient data ferries between clusters due to their high mobility. This requires up-to-date knowledge about cluster locations to determine UAV flight paths. However, the shortest paths usually do not cover disconnected network nodes in transit between clusters, that will miss critical messages like evacuation notices or hazard warnings.This paper provides two contributions for UAV-assisted post-disaster DTN communication. First, we present a novel approach to estimate dynamically changing cluster locations in a post-disaster scenario. Second, we introduce a topology-aware path planning approach for UAV data ferry flights, covering in-transit nodes in-between clusters. Our evaluation highlights the requirements on network topology knowledge for an efficient application of UAV data ferries. We furthermore demonstrate that our approach significantly reduces the number of disconnected in-transit nodes, which is especially important in the considered post-disaster scenario.