Course Objectives

This course introduces participants to the integration of Artificial Intelligence (AI) and Machine Learning (ML) techniques in the monitoring, optimization, and enhancement of Radio Access Network (RAN) performance. The focus is on practical applications of AI/ML-driven insights for KPI improvement, automation, and root cause analysis within 4G and 5G network environments.

Participants will gain hands-on understanding of how AI/ML can automate optimization workflows, improve troubleshooting accuracy, and enhance the efficiency of radio planning and operations.

Contents

• Introduction to RAN Performance Optimization
  • Key RAN KPIs: Accessibility, retainability, mobility, throughput
  • Overview of classical vs AI-driven optimization approaches
• AI/ML Fundamentals in the Context of Telecom
  • Basics of machine learning for engineers
  • Supervised vs unsupervised learning for performance analytics
• Data Sources and Preprocessing
  • KPI datasets, OSS, drive tests, probe data
  • Data cleansing, feature engineering, and model input structuring
• Use Cases of AI/ML in RAN Optimization
  • Anomaly detection in KPIs
  • Traffic forecasting and capacity planning
    • Clustering techniques for cell/devices performance classification
    • Coverage hole / interference spot identification
    • Smart Energy saving
  • Mobility event optimization using AI
  • Automated parameter tuning (e.g., HO, reselection, power control)
• Automation Frameworks and Tools
  • Overview of AI/ML tools applied to RAN
  • Role of OSS, third-party analytics, and custom ML pipelines
• Performance Impact Assessment
  • Model evaluation metrics and field validation techniques
  • Integration with existing network monitoring workflows
• LLMs for Root cause Analysis
  • How LLMs can speed up RCA in 4G/5G RAN by analyzing KPIs, counters, logs and configuration context together
  • How to use prompt templates to produce ranked root-cause hypotheses with evidence and next validation steps
  • How to apply guardrails (RAG, rules, confidence scoring, human review) for reliable, engineer-ready outputs
• Future Trends and Next Steps
  • Towards cognitive RAN and self-optimizing networks (SON)
  • Role of AI in Open RAN (O-RAN) architectures

Target Outcomes

Participants will be equipped with the competence to:

• Understand and evaluate AI/ML techniques applicable to RAN optimization
• Apply data-driven methods to improve KPI performance
• Support the evolution toward automated and cognitive network operations

Entry Requirements

A foundational understanding of mobile communication systems, including 4G (LTE) and 5G technology.

Experience in radio network design and optimization

Knowledge of RF concepts and principles

Target Group

Radio Planning and Optimization Engineers

RAN Performance Analysts

Technical Professionals working with network analytics and automation

Mode of Delivery: Instructor-led learning in a physical virtual classroom

Duration: The duration of this course is three or four days