Related Experiment Video
Updated: Jul 15, 2026
05:49
Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
A survey of the integration between machine learning and artificial intelligence techniques in software-defined
Ariel Łukowski1, Grzegorz Papier1, Robert Wójcik1
1Institute of Telecommunications and Cybersecurity, AGH University of Krakow, al. Mickiewicza 30, Krakow, 30-059, Poland.
Summary
Software-Defined Networking (SDN) and Artificial Intelligence (AI) integration is key for intelligent network management. This survey maps AI techniques across SDN domains, highlighting trends and gaps for future autonomous networks.
Area of Science:
- Computer Science
- Telecommunications Engineering
- Artificial Intelligence
Background:
- Software-Defined Networking (SDN) offers programmable and flexible network control.
- SDN facilitates integration with technologies like NFV, Edge/Fog/Cloud Computing, and 5G/6G.
- AI techniques are increasingly vital for network management tasks such as anomaly detection and traffic analysis.
Purpose of the Study:
- To comprehensively survey the integration of SDN and AI across diverse networking domains.
- To analyze and classify over 400 publications from 2015-2025 based on network domain, AI application, and AI learning methods.
- To develop a mapping framework illustrating the relationships between AI techniques and SDN application domains.
Main Methods:
- Systematic literature review and analysis of over 400 publications (2015-2025).
- Classification of research based on network domain (core, cloud, edge, wireless, IoT, vehicular), AI application area, and AI learning paradigms (supervised, unsupervised, hybrid).
- Development of a taxonomy-driven mapping framework to visualize AI-SDN integration trends and identify research gaps.
Main Results:
- The integration of SDN and AI forms the foundation for intelligent network management.
- AI techniques, particularly supervised, unsupervised, and hybrid learning, support prediction, classification, anomaly detection, and optimization in SDN.
- Current research predominantly focuses on these AI paradigms, with gaps identified in areas like interactive learning and closed-loop control for full automation.
Conclusions:
- SDN and AI integration is crucial for advancing network management towards greater automation and intelligence.
- Future research should focus on developing more autonomous, adaptive, and explainable network management architectures.
- Addressing gaps in interactive learning and multi-domain orchestration is essential for achieving fully automated networks.