Navigation Data-Assisted Opportunistic Spectrum Scheduling for Network-Based UAV Systems: A Parallel Restless Bandits Formulation

Abstract

A myriad of applications and services of information and communication technology (ICT) has been developed in the past decades, within which networks are the fundamental architecture to connect people and things. As one of the network-based ICT applications, recently, unmanned aerial vehicle (UAV) cooperation systems have attached much attention due to their successful applications in complex military and civilian missions. In this paper, we propose a navigation data-assisted optimal opportunistic spectrum access scheme for wireless communications in heterogeneous UAV networks, to achieve maximized long-term data rate by flexibly scheduling the spectrum subbands. Thanks to the navigation data that is always available locally at the entities in the network, rough prediction of wireless link quality can be obtained to assist subband allocations. Furthermore, to formulate the optimal spectrum scheduling problem, we develop a parallel restless bandits model, which can reduce the NP-hard optimization problem to simply selecting the link with the minimum index in the online stage. Extensive simulation results are also presented to demonstrate the significant performance improvement of the proposed scheme compared to the existing one.