Energy Efficiency Optimization for RIS-Assisted UAV-Enabled MEC Systems

Abstract

The reconfigurable intelligent surface (RIS) can proactively modify the wireless communication environment and further improve the service quality of the wireless networks. Motivated by this vision, in this paper, we propose to introduce the RIS into the unmanned aerial vehicle (UAV) enabled mobile edge computing (MEC) systems. Considering both the amount of completed task bits and the energy consumption, the energy efficiency of the RIS-assisted UAV-enabled MEC systems is maximized by jointly optimizing the bit allocation, phase shift, and UAV trajectory via an iterative algorithm with a double-loop structure. Simulation results show that: 1) the UAV tends to fly closer to the RIS rather than the IoT devices; 2) the energy efficiency first increases and then decreases with the increase of the total amount of task-input bits of IoT devices; 3) higher energy efficiency can be achieved by our proposed algorithm.

Resource and trajectory optimization for secure communication in RIS assisted UAV‐MEC system

“…Due to the height of the UAV and RIS, the channel state of the UAV-RIS is assumed to be a link-of-sight (LoS) channel, then h ru [n] and h re [n] can be expressed as follows, respectively [36].…”

“…At time slot n , the channels from user m to the UAV_L, user m to the RIS, RIS to the UAV_L, GJ to the RIS, user m to the UAV_E, RIS to the UAV_E, and GJ to the UAV_E are represented as $$h_{mu}[n]\in \mathbb {C}$$, $${\mathbf {h}_{mr}\in \mathbb {C}^{K\times 1}}$$, $${\mathbf {h}_{ru}\in \mathbb {C}^{K\times 1}}$$, $${\mathbf {h}_{Jr}[n]\in \mathbb {C}^{K\times 1}}$$, $${h_{me}[n]\in \mathbb {C}}$$, $${\mathbf {h}_{re}\in \mathbb {C}^{K\times 1}}$$ and $${h_{Je}[n]\in \mathbb {C}}$$, respectively. Due to the height of the UAV and RIS, the channel state of the UAV‐RIS is assumed to be a link‐of‐sight (LoS) channel, then $$\mathbf {h}_{ru}[n]$$ and $$\mathbf {h}_{re}[n]$$ can be expressed as follows, respectively [36]. …”

“…The ground‐to‐RIS channel is modeled as a Rician fading channel due to the complex electromagnetic environment caused by reflections from buildings [36]. The user‐RIS and jammer‐RIS channel models are formulated as follows, respectively.…”

“…[n] denotes the transmit power of the mth user in the l th iteration at the nth time slot. Similarly, applying a firstorder Taylor expansion for s 2,m [n] at point p (l ) m [n], (34 c) can be transformed as(36), shown at the bottom of the next page.…”