A learning-based expected best offloading strategy in wireless edge networks

Yi Chen Wu, Thinh Quang DInh, Yaru Fu, Che Lin, Tony Q.S. Quek

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Citations (Scopus)

Abstract

Recently, Mobile-Edge Computing (MEC) has been considered as a powerful supplement to a wireless network by processing computationally intensive tasks for resource-limited mobile devices. However, despite saving computational energy at User Equipment (UE), there is additional transmission energy consumption. As a result, the joint offloading strategy should be carefully selected to save energy and computational time. In this work, we investigated a sum cost minimization problem in a multi-UE multi-computing access point (CAP) system with time-varying channels. Our approach combines the optimization-based resource allocation algorithm with a Q-learning-based strategy selection mechanism. Without the need for communication overhead for CSI and inter- neighborhood cost value exchange, our algorithm shows prominent performance over the benchmark schemes with moderate assumptions.

Original languageEnglish
Title of host publication2019 IEEE Global Communications Conference, GLOBECOM 2019 - Proceedings
ISBN (Electronic)9781728109626
DOIs
Publication statusPublished - Dec 2019
Externally publishedYes
Event2019 IEEE Global Communications Conference, GLOBECOM 2019 - Waikoloa, United States
Duration: 9 Dec 201913 Dec 2019

Publication series

Name2019 IEEE Global Communications Conference, GLOBECOM 2019 - Proceedings

Conference

Conference2019 IEEE Global Communications Conference, GLOBECOM 2019
Country/TerritoryUnited States
CityWaikoloa
Period9/12/1913/12/19

Keywords

  • Energy consumption
  • Latency
  • Mobile edge computing
  • Non-convex optimization
  • Offloading
  • Q-learning
  • Resource management

Fingerprint

Dive into the research topics of 'A learning-based expected best offloading strategy in wireless edge networks'. Together they form a unique fingerprint.

Cite this