logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

Energy-efficient offloading for bidirectional USV computation tasks in DT-supported RIS-assisted UAV-USV MEC network

delete2026-07-07
delete0
PRE
AI
C
Chao Ma
Q
Quan Liu
Y
Yangzhe Liao *
DOI：10.1016/j.adhoc.2026.104342delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
The integration of digital twin (DT) technology with multi-access edge computing (MEC) networks establishes a novel paradigm enabling practical bidirectional data computation for unmanned surface vehicles (USVs). The scenario may become even more optimistic with the implementation of unmanned aerial vehicle (UAV)-carried reconfigurable intelligent surface (RIS) transmission scheme to serve wireless communications between USVs and terrestrial base stations (TBS), where they generally suffer severe blockage from harsh inland waterway environments. In this work, a DT-supported MEC network architecture is proposed, where tethered UAV-carried RIS-assisted data transmission scheme is utilized to assist USVs bidirectional data computation. To address the energy minimization challenge of USVs, we formulate an optimization framework that jointly considers USVs bidirectional computation tasks execution decisions, TUAVs hovering coordinates, DT estimated USVs computing capabilities and transmission powers. A heuristic solution is introduced that partitions the formulated problem into two subproblems, e.g., the joint optimization of USVs bidirectional computation task execution decisions, DT estimated USVs computing capabilities and transmission powers subproblem, and the optimization of TUAVs hovering coordinates subproblem. Following this, each subproblem is effectively resolved by the proposed Hungarian fusion differential evolution (HFDE) algorithm and enhanced particle swarm optimization (EPSO) algorithm, respectively. The proposed solution can effectively decrease about 40% USVs energy consumption compared with benchmarks. Moreover, successfully executed USVs tasks versus the maximum allowable latency are investigated.
Keywords:
Unmanned surface vehicle
Unmanned aerial vehicle
Multi-access edge computing
Digital twin

Journal

Ad Hoc Networks cover
Ad Hoc Networks
IF:
4.8
Papers:
481
Citations:
6.2K

Organization

No organization information available