Joint Obstacle Avoidance and 3D Deployment for Securing UAV-Enabled Cellular Communications

Dong Wang,Yanping Yang

doi:10.1109/access.2020.2986026

Abstract

Unmanned aerial vehicle (UAV) can enable cellular mobile communications in unexpected or temporary scenarios, but there are information security problems raised in such applications. Physical layer security is now emerging as a promising alternative technology to realize secrecy in wireless communications. In the UAV-enabled communication systems, the fully controllable mobility of UAV provides new design degrees of freedom for enhancing physical layer security. This paper investigates an UAV-enabled cellular communication system where an ultralow-altitude UAV is equipped with a base station to serve a ground user in the presence of an eavesdropper. The probability of non-zero secrecy capacity is maximized subject to the airspace and obstacle constraints by adaptively optimizing the UAV's three-dimensional positions. The optimization problem is simplified by equivalent transformation and then solved by semidefinite relaxation (SDR). Two approaches termed as eigenvector approximation and random sampling are addressed to extract an approximation solution for the primal problem from the solution obtained by the SDR. Simulation results show that there is a slight difference between the probability of non-zero secrecy capacity achieved by the two solution approximation approaches.

Highlights

Unmanned aerial vehicle (UAV) is rapidly growing in a wide range of the fifth generation (5G) and beyond networks
The probability of non-zero secrecy capacity achieved by the random sampling approach is just a little lower than that achieved by the exhaustive search, but outperforms that achieved by the eigenvector approximation approach
We can see that the UAV intelligently changes its spatial positions according to the movement of the eavesdropper to increase the probability of non-zero secrecy capacity as much as possible

Summary

INTRODUCTION

Unmanned aerial vehicle (UAV) is rapidly growing in a wide range of the fifth generation (5G) and beyond networks. UAV can be applied into cellular communications to support some commercial applications which are either unexpected or temporary, such as providing emergency communication service after a natural disaster, offloading data of extremely dense users in hotspot areas, or establishing communication connections in complex terrain [2], [3] In this context, many works focus on trajectory optimization to improve the performance of UAV communication networks. In [9], [17], UAV-enabled mobile jamming schemes to improve the secrecy rate are addressed by optimizing the UAV’s trajectory, jointly with its communicating/jamming power allocations. We use the probability of non-zero secrecy capacity as a performance metric to investigate the physical-layer security transmission in the UAV-enabled cellular communications.

SYSTEM MODEL

SOLUTION APPROXIMATION

SIMULATION RESULTS

CONCLUSION