Enhancing 5G Network Slicing: Slice Isolation Via Actor-Critic Reinforcement Learning with Optimal Graph Features
Javadpour, Amir; Ja'fari, Forough; Taleb, Tarik; Benzaïd, Chafika (2024-02-26)
IEEE
26.02.2024
A. Javadpour, F. Ja'fari, T. Taleb and C. Benzaïd, "Enhancing 5G Network Slicing: Slice Isolation Via Actor-Critic Reinforcement Learning with Optimal Graph Features," GLOBECOM 2023 - 2023 IEEE Global Communications Conference, Kuala Lumpur, Malaysia, 2023, pp. 31-37, doi: 10.1109/GLOBECOM54140.2023.10437687
https://rightsstatements.org/vocab/InC/1.0/
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
https://rightsstatements.org/vocab/InC/1.0/
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
https://rightsstatements.org/vocab/InC/1.0/
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:oulu-202402292046
https://urn.fi/URN:NBN:fi:oulu-202402292046
Tiivistelmä
Abstract
Network slicing within 5G networks encounters two significant challenges: catering to a maximum number of requests while ensuring slice isolation. To address these challenges, we present an innovative actor-critic Reinforcement Learning (RL) model named ‘Slice Isolation based on RL’ (SIRL). This model employs five optimal graph features to construct the problem environment, the structure of which is adapted using a ranking scheme. This scheme effectively reduces feature dimensionality and enhances learning performance. SIRL was assessed through a comparative analysis with nine state-of-the-art RL models, utilizing four evaluation metrics. The average results demonstrate that SIRL outperforms other models with a 70% higher coverage rate of requests and an 8% reduction in damage resulting from DoS/DDoS attacks.
Network slicing within 5G networks encounters two significant challenges: catering to a maximum number of requests while ensuring slice isolation. To address these challenges, we present an innovative actor-critic Reinforcement Learning (RL) model named ‘Slice Isolation based on RL’ (SIRL). This model employs five optimal graph features to construct the problem environment, the structure of which is adapted using a ranking scheme. This scheme effectively reduces feature dimensionality and enhances learning performance. SIRL was assessed through a comparative analysis with nine state-of-the-art RL models, utilizing four evaluation metrics. The average results demonstrate that SIRL outperforms other models with a 70% higher coverage rate of requests and an 8% reduction in damage resulting from DoS/DDoS attacks.
Kokoelmat
- Avoin saatavuus [38865]