The demands of 5G infrastructure mean CSPs face rising energy costs, as well as pressure from regulators to reduce their environmental impact. Without effective energy management, CSPs risk missing both their financial and sustainability targets. The radio access network (RAN) accounts for nearly 70% of total network energy consumption, making it a critical area for operational and environmental cost management.
This Catalyst is exploring how AI-driven, autonomous solutions can continuously learn, predict and address the RAN’s energy-requirements in real-time. The team is developing an AI-based energy management system that forecasts and adjusts RAN power levels based on real-time traffic data, energy consumption data, environmental factors and network topologies. During low-traffic periods, the solution can automatically adjust RAN capacity and RAN components’ energy consumption, entering them into low-power states without affecting the network performance.
Aligned with TM Forum’s Green Network standards, the solution is designed to adhere to best practices in AI governance and sustainability. By harnessing TM Forum Open APIs for data exchange and AI models for decision-making, this project demonstrates how CSPs can leverage standards-based automation to reduce operational overheads and enhance environmental impact.
The Catalyst is aiming for a 15% reduction in RAN energy consumption and associated costs within the first 12 months of deployment. It is also pursuing a 15-20% reduction in carbon footprint for CSPs’ RAN operations. The project team estimate that even a 10% reduction in energy consumption could collectively save CSPs’ globally US$1.8 billion a year. Assuming the cost of implementing the solution is about US$2 million, the solution could yield a very high return on investment.
As well as addressing immediate business pain points, a comprehensive, sustainable approach to RAN management will make networks more reliable, eco-friendly and cost-effective, positioning CSPs to thrive in a 5G and AI-driven future.