MODELING A UAV-BASED MESH NETWORK TO ANALYZE LATENCY AND THROUGHPUT

Abstract

As the military has moved toward network-centric operations, unmanned aerial vehicles (UAVs) have become increasingly valuable for capturing real-time information for joint operations on the ground. These UAVs contribute to the mission by conveying the scene accurately to the decision maker. However, because of constraints in free space, using MANET with UAV nodes in free space that has no pre-existing infrastructure poses a network latency problem. Latency reduces network efficiency when transmitting real-time information to a ground station. One goal of analyzing latency is understanding the logic of the UAV network architecture and protocols. Specifically, the thesis investigates the following question: What are the effects of UAV altitude on MANET latency? To investigate this question, this research simulates a two-node UAV mesh network at 2,000 and 3,000 feet altitude using Riverbed Modeler Academic Edition software. The research finds that there is no significant difference in latency at these altitudes. Therefore, all else being equal, 3,000 feet is the preferred altitude for UAVs, as it allows observation of wider areas and reduces UAV vulnerability to enemy weapons without affecting latency.