Loading web-font TeX/Main/Regular
Towards Energy-And Cost-Efficient Sustainable MEC-Assisted Healthcare Systems | IEEE Journals & Magazine | IEEE Xplore

Towards Energy-And Cost-Efficient Sustainable MEC-Assisted Healthcare Systems


Abstract:

To meet the demands of new healthcare applications with high computation complexity, multi-access edge computing (MEC) is emerging as a key component of modern healthcare...Show More

Abstract:

To meet the demands of new healthcare applications with high computation complexity, multi-access edge computing (MEC) is emerging as a key component of modern healthcare systems which provides rich computing services to the users. With the increase in the number of wireless body area network (WBAN) users requesting computing services, the computational load on the MEC servers increases. A major issue related to the operation of these MEC servers is their sustainability in terms of energy consumption and heavy carbon emission. Therefore, in this work, we propose a resource management scheme, which minimizes the energy consumption of the MEC server without compromising on the quality-of-experience (QoE) of the WBAN users. For that, we propose a cooperative framework between the MEC server and WBAN users, where the MEC server motivates WBAN users to opt for partial offloading instead of full offloading of the computing services. More specifically, the MEC server bargains with each WBAN user for the amount of the task it compute locally and the corresponding reimbursement. We model this economic interaction between the MEC server and all the participating WBAN users using the Nash bargaining theory. Thereafter, we derive the closed-form Nash bargaining solutions (NBS) for two different bargaining protocols. Finally, numerical results show the proposed bargaining scheme is capable of improving the MEC server payoff by 44.3\%, 51.4\%, and 56.1\%, respectively, compared to the state-of-the art schemes.
Published in: IEEE Transactions on Sustainable Computing ( Volume: 7, Issue: 4, 01 Oct.-Dec. 2022)
Page(s): 958 - 969
Date of Publication: 26 April 2022

ISSN Information:


Contact IEEE to Subscribe

References

References is not available for this document.