Project Overview
The IBN@TEIN project represents a groundbreaking initiative to establish an AI-driven intent-based networking platform across multiple Asian countries. By integrating Software-Defined Networking (SDN), Machine Learning (ML), and real-time monitoring into a closed-loop control system, this project automates service deployment with Quality of Service (QoS) assurance across the TEIN infrastructure.
Participating Countries
Jeju National University, NIA
PERN, NUST, SUIT
University of Malaya, MYREN
Institute of Technology, CamREN
Project Objectives
AI-Enabled SDN Architecture
Deploy domain-based intelligent J-boxes with advanced SDN capabilities for automated network management and service orchestration.
ML-Driven Predictive Routing
Train machine learning models using monitored network metrics to enable intelligent path selection and predictive network optimization.
Closed-Loop IBN System
Establish a comprehensive intent-based networking system capable of managing resources in real-time with minimal human intervention.
Regional Collaboration
Facilitate hands-on training and knowledge transfer for federated deployment across participating countries and institutions.
System Architecture
The IBN@TEIN platform leverages cutting-edge technologies to create an intelligent, self-managing network infrastructure that adapts to changing conditions and requirements in real-time.
Key Activities & Implementation
Feasibility & Training Workshops
Conducted comprehensive feasibility meetings and online training workshops focusing on J-box deployment and SDN/IBN components across all participating countries.
SDN Cloud Site Deployment
Successfully deployed three SDN-enabled cloud sites in Malaysia and Korea, establishing the foundational infrastructure for the IBN platform.
Monitoring Infrastructure
Implemented real-time monitoring capabilities using Grafana and Prometheus on deployed instances, providing comprehensive network visibility and analytics.
ML Model Development
Developed and integrated ML-driven path selection models with SDN controllers, enabling intelligent routing decisions based on network conditions and performance metrics.
APAN54 Participation
Presented project outcomes at APAN54 conference and conducted specialized training sessions on PySpark and IBN system implementation for regional stakeholders.
Key Outcomes & Achievements
Technical Achievements
- AI-Powered IBN Architecture: Successfully prototyped and deployed an intelligent intent-based networking architecture in laboratory and test environments, demonstrating autonomous network management capabilities.
- Intelligent J-box Network: Established intelligent J-boxes at 3 partner sites with comprehensive monitoring modules, creating a distributed intelligence network across the region.
- Comprehensive Monitoring Stack: Integrated advanced monitoring tools including Grafana, Node Exporter, and Prometheus for real-time network visibility and performance analytics.
- Regional Capacity Building: Trained stakeholders through hybrid and physical workshops, APAN meetings, and specialized technical sessions, building regional expertise in IBN technologies.
Innovation Highlights
Predictive Path Selection
ML algorithms analyze network metrics to predict optimal routing paths, reducing latency and improving overall network performance through intelligent decision-making.
Autonomous Service Deployment
Intent-based networking capabilities enable automatic service deployment and configuration based on high-level policies and requirements.
Multi-Domain Federation
Federated SDN architecture allows seamless collaboration and resource sharing across different administrative domains and countries.
Real-time Adaptation
Closed-loop control system continuously monitors network conditions and automatically adapts configurations to maintain optimal performance.
Challenges & Strategic Solutions
Challenges Addressed
Strategic Solutions
Contributing to Sustainable Development Goals
Future Directions & Roadmap
Next Phase Deployment
The project roadmap includes expanding IBN capabilities to Cambodia and Pakistan following equipment delivery, with comprehensive monitoring infrastructure to support full ML model training and deployment across all participating countries.
Enhanced Autonomy
Future development will focus on enabling fully autonomous orchestration via the IBN engine with enhanced multi-domain support, allowing for seamless service deployment and management across the entire TEIN network infrastructure.
Vision for Intelligent Networks
The IBN@TEIN project is pioneering the future of intelligent networking in Asia, where AI-driven systems will autonomously manage complex network infrastructures, optimize performance in real-time, and enable unprecedented levels of service quality and reliability. This foundation will support the next generation of research, education, and innovation across the region.
Scaling & Replication
Regional Expansion
Extend IBN capabilities to additional TEIN member countries, creating a comprehensive intelligent network infrastructure across Asia-Pacific.
Enhanced ML Models
Develop more sophisticated machine learning algorithms incorporating advanced analytics, predictive capabilities, and autonomous decision-making processes.
Cross-Domain Integration
Implement seamless integration between different network domains, enabling unified management and optimization across diverse infrastructure environments.
Industry Standards
Contribute to the development of industry standards for intent-based networking and AI-driven network management in research and education networks.