Resource Request Dispatch in Standalone and Federated MEC Systems: A Matching Game Approach

Ming Yi Lin, Li-Hsing Yen, Hojjat Baghban

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

Abstract

Multi-access edge computing (MEC) system consisting of geographically-distributed heterogeneous servers can provide low-latency virtualized resource to support computation offloading of smart devices. When bulk offloading requests comes to an MEC system, how to dispatch requests to servers so as to maximize divergent objectives of MEC service providers and users is challenging. The problem further involves money transfer when different MEC service providers can share resource to each other. In this paper, we address request dispatch issues in a standalone MEC and among federated MEC systems using matching game theory. We have adapted several classical matching algorithms to our problem. Simulation results show that we can serve more requests while still meeting latency constraints. For federated MEC systems, we can also have high revenue.

Original languageEnglish
Title of host publication2019 IEEE Wireless Communications and Networking Conference, WCNC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538676462
DOIs
StatePublished - Apr 2019
Event2019 IEEE Wireless Communications and Networking Conference, WCNC 2019 - Marrakesh, Morocco
Duration: 15 Apr 201919 Apr 2019

Publication series

NameIEEE Wireless Communications and Networking Conference, WCNC
Volume2019-April
ISSN (Print)1525-3511

Conference

Conference2019 IEEE Wireless Communications and Networking Conference, WCNC 2019
CountryMorocco
CityMarrakesh
Period15/04/1919/04/19

Keywords

  • Edge Computing
  • Matching Theory
  • Resource Allocation

Fingerprint Dive into the research topics of 'Resource Request Dispatch in Standalone and Federated MEC Systems: A Matching Game Approach'. Together they form a unique fingerprint.

Cite this