Placement of UAVs as communication relays aiding mobile ad hoc wireless networks

Unmanned aerial vehicles (UAVs) have been used for surveillance and reconnaissance operations. Such communications enabled platforms can also be effectively utilized to enhance the communications transport capabilities of a mobile ad hoc wireless network. When properly embedded into the architecture of a communications network, the resulting UAV aided terrestrial network architecture is expended into a multi-layered hierarchical network structure. The UAV aided network can provide for transport of flows that span longer distances, yield better reliability, higher mobility based robustness and upgraded throughput capacity. We have recently proposed the 'robust throughput' and 'robust throughput capacity' measures to characterize the capability of a mobile ad hoc wireless network to provide for robust and survivable transport of flows. Robust service is critically required for supporting applications that involve flow transactions that must not, with high probability, be transported along routes that are prematurely interrupted. To enhance the robust throughput performance of mobile ad hoc wireless networks, we present in this paper a method and algorithm that are used to place relay nodes, such as UAVs, in locations that efficiently serve to support the robustness and capacity requirements of the underlying mobile ad hoc wireless network system, and to compute the optimal (flow admission oriented) regulation and distribution of traffic flow classes across terrestrial and UAV based routes. Our schemes are used to determine effective 3-D coordinates for placing a UAV relay node to provide for such joint performance upgrade. We employ performance metrics that ensure that network capacity resources are allocated to flow classes in a 'fair' manner. In doing so, different classes can be attached distinct weight measures or utility functions, as well as specify minimum and maximum desired capacity levels. Traffic flows that are generated within each flow class can be guided along an identified terrestrial route, be directed to traverse a UAV based path, or be optimally split between terrestrial and UAV routes. We present a mathematical method for the optimal joint calculation, for each class of flows, of the best traffic rate split and of the maximum flow admission rate threshold level. We present an UAV aided network configuration to illustrate the characteristics of our placement algorithm.