Maximizing Throughput of UAV-Relaying Networks with the Load-Carry-and-Deliver Paradigm

We consider the task of using one or more unmanned aerial vehicles (UAVs) to relay messages between two distant ground nodes. For delay-tolerant applications like latency-insensitive bulk data transfer, we seek to maximize throughput by having a UAV load from a source ground node, carry the data while flying to the destination, and finally deliver the data to a destination ground node. We term this the "load-carry-and-deliver" (LCAD) paradigm and compare it against the conventional multi-hop, store-and-forward paradigm. We identify and analyze several of the most important factors in constructing a throughput-maximizing framework subject to constraints on both application allowable delay and UAV maneuverability. We report performance measurement results for IEEE 802.11g devices in three flight tests, based on which we derive a statistical model for predicting throughput performance for LCAD. Due to the nature of commercial off-the-shelf systems, this methodology is of essential importance for allowing better flight-path design to achieve high throughput.