Detailed Information

Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

Intelligent Offloading and Resource Allocation in HAP-Assisted MEC Networks

Authors
Lakew, Demeke Shumeye.Tran, Anh-TienDao, Nhu-NgocCho, Sungrae
Issue Date
Dec-2021
Publisher
IEEE Computer Society
Keywords
aerial access network; Deep reinforcement learning; HAP; MEC; resource allocation; task offtoading
Citation
International Conference on ICT Convergence, v.2021, no.10, pp 1582 - 1587
Pages
6
Journal Title
International Conference on ICT Convergence
Volume
2021
Number
10
Start Page
1582
End Page
1587
URI
https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/54909
DOI
10.1109/ICTC52510.2021.9621158
ISSN
2162-1233
Abstract
An aerial platform such as high altitude platform (HAP) is emerging as a promising technology to enhance the capacity, coverage, and computation experience of user devices (UDs) in fifth (5G) and beyond (B5G) generation wireless networks, especially in underserved areas that have no coverage of ground base stations (GBSs). In particular, a HAP equipped with a computing server can provide computation and communication resources to resource-constrained UDs for computing their tasks with various delay requirements on demand. Thus, in this paper, we study the partial task offloading and communication resource allocation in HAP-assisted edge computing and formulated the problem to maximize the total number of accomplished tasks of UDs with satisfied delay requirements while minimizing their total energy consumption. To make a real-time decision while considering the network dynamics and heterogeneous task requirements, we transform the problem into an Markov decision process (MDP)-based problem and introduce an algorithm based on the deep deterministic policy gradient (DDPG), named DDPG-PORA. We carry out simulation experiments and the results show that DDPG-PORA reliably converges and provides higher performance than the comparison methods. © 2021 IEEE.
Files in This Item
There are no files associated with this item.
Appears in
Collections
College of Software > School of Computer Science and Engineering > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher Cho, Sung Rae photo

Cho, Sung Rae
소프트웨어대학 (소프트웨어학부)
Read more

Altmetrics

Total Views & Downloads

BROWSE