An Online Learning Algorithm for Distributed Task Offloading in Multi-Access Edge Computing
- Submitting institution
-
King's College London
- Unit of assessment
- 12 - Engineering
- Output identifier
- 130052690
- Type
- D - Journal article
- DOI
-
10.1109/TSP.2020.2991383
- Title of journal
- IEEE Transactions on Signal Processing
- Article number
- 9082169
- First page
- 3090
- Volume
- 68
- Issue
- -
- ISSN
- 1053-587X
- Open access status
- Compliant
- Month of publication
- April
- Year of publication
- 2020
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
1
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper introduces a novel energy-efficient algorithm that enables battery-powered smart devices to perform computation-intensive tasks in real-time at the edge of cellular networks. The algorithm is a potential key enabler of ultra-low latency applications in 5G and 6G networks. Autonomic online decision-making by smart devices in time-varying wireless environment is the significant feature of the proposed distributed algorithm, which is enabled by an innovative integration of statistical bandit iterations in optimisation theory. The bandit optimisation, introduced in this paper for the first time, extends the myopic optimisation to handle foresighted functionalities via the online-learning capability of statistical bandit theory.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -