Joint Optimization of Traffic and Computation Offloading in UAV-Assisted Wireless Networks

Abstract

Wireless communication via unmanned aerial vehicles (UAVs) is a promising way to provide transmission coverage and computation capacity for mobile user devices, especially in remote areas where the communication resources and infrastructures are extremely limited. In this paper, we study a UAV-assisted traffic and computation offloading (UTCO) problem, where one UAV can provide communication and computation capabilities for its covered user equipments (UEs). The UTCO problem is formulated as a problem of maximizing user satisfaction, in which the bandwidth allocation, transmit power including the UEs and UAV, and UAV trajectory are jointly considered. However, the UTCO problem is proven to be non-convex and some variables are nonlinear coupled, which cause it difficult to be solved optimally. To tackle this challenge, we propose a two-stage alternative optimization approach by leveraging successive convex approximation (SCA) method to obtain a near-optimal solution. The simulation results show that the proposed approach can achieve an outstanding performance in convergence speed and user satisfaction.