Resource Allocation and Trajectory Design for MISO UAV-Assisted MEC Networks

Mobile Edge Computing (MEC) is a promising technology in the next generation network, which provides computing services for user equipments (UEs) to reduce the task delay and prolong the usage time of UEs. To address the deficiency of poor channel quality caused by multipath and blockages in traditional MEC networks, a multiple input single output (MISO) UAV-assisted MEC network is studied. A system energy consumption minimization problem is formulated by jointly optimizing the the UAV’s beamforming vectors, the UAV’s central processing unit (CPU) frequency, the UAV’s trajectory, the UEs’ transmission power and the UEs’ CPU frequency subject to the constraints on the task, the UAV’s trajectory, and the UEs’ computation tasks. A three-stage iterative algorithm is proposed to solve the challenging non-convex problem. The closed-form expressions for the optimal UAV CPU frequency and the transmission power of UEs are derived. Simulation results show that the proposed algorithm is superior to the benchmark schemes in terms of energy consumption, and the convergence performance is guaranteed.

I. Introduction

With the rapid development of 5G and the Internet of Things (IoT) technology, massive smart devices need to access the network and the amount of task data in the network reaches an unprecedented size [1]. Moreover, diverse emerging sophisticated Internet applications are emerging, such as online gaming, virtual reality, augmented reality, etc., most of which have strict requirements for latency, security, and other metrics [2]. In order to provide high quality service to user equipments (UEs) and guarantee their quality of service (QoS) requirements, it is necessary to perform a large number of computing tasks on wireless devices in a short duration. However, UEs usually have a limited battery capacity and limited computation capability, making it challenging to handle these computation-intensive and latency-sensitive tasks locally. Although cloud computing can offload the computing tasks of UEs to cloud servers to relieve the pressure of mobile devices, in the case of IoT, offloading massive computing data to cloud servers can result in core network congestion and excessive delay issue.