Joint resource allocation and trajectory design for secure UAV-enabled communications

Abstract

Currently, the unmanned aerial vehicle (UAV)-enabled communications are emerging solutions to support the conventional terrestrial network and cellular communications such as enhancing coverage of ground base stations, offloading the data traffic, and so on. UAVs can be also exploited in various military and civil applications (e.g. battlefield, disaster, and hazardous region). In particular, exploiting the UAVs as temporary base stations have many advantages such as dominant line-of-sight air-to-ground channel, flexible deployment by on-demand requirements of ground nodes, and mobility compared to the quasi-stationary terrestrial base station. By exploiting the characteristics of these UAVs, many types of UAV-enabled communications are studied such as UAV transmitters, UAV jammers, and UAV relays. Moreover, the physical layer security of UAV-enabled communications can be significantly enhanced through the characteristic of UAVs by proper trajectory design and resource allocation. UAV jammers and UAV relays have been investigated for enhancing physical layer security given perfect location information of ground nodes including eavesdroppers. However, the availability of the location information of ground nodes is limited by the covert nature of Eves and the estimation errors of the global navigation satellite system (GNSS).

In this dissertation, we investigate the physical layer security of two types of UAV-enabled communications: UAV jammer scenario and dual UAV scenario (i.e., UAV relay and UAV jammer). Besides, we assume that the availability of the location information of ground nodes is limited by GNSS jamming/spoofing and practical estimation errors of GNSS and wireless localization. To tackle this issue, considering the location uncertainty of ground nodes, the robust trajectory design and resource allocation of UAV jammer are proposed for multiple ground nodes including multiple eavesdroppers. In the UAV jammer scenario, we aim to maximize the minimum secrecy rate among the multiple Bobs in the worst-case scenario of ground nodes. Moreover, we consider the propulsion energy constraint of UAV jammer. In the dual UAV scenario, we consider two optimization problems: secrecy rate maximization and secrecy energy efficiency maximization. To handle the non-convexity of the optimization problems, we propose an iterative algorithm based on the successive convex approximation and block coordinate descent scheme. Numerical results verify the performance of the proposed algorithm compared to benchmark schemes and provide insight to establish secure communications.