Smart coverage adjustment between CSPs brings us a step closer to carbon net zero

Commercial context

As network infrastructure OPEX and CAPEX increases, CSPs are deepening their collaboration and sharing resources to increase efficiency, improve automated processes (such as provisioning) and increase QoS and QoE. As the industry seeks to fulfil its objective of carbon neutrality, there is renewed motivation to work together to reduce consumption and energy usage. One way of hitting this target is by ensuring coverage is provided in the most efficient way possible.

For this to be possible, the industry needs a reliable automated framework for communication between CSPs to align responsibilities for maintaining consistent coverage, regardless of who the subscriber belongs to. The resulting energy savings are significant, but can only be achieved when commercially viable, and when the regulators permit switching off an entire RAN. Operationally, the main challenge is how to reconfigure the network to adapt its coverage areas in compliance with the CSPs' policies. Participating CSPs face an additional challenge in forming a coordinated and consistent inter-CSP accounting and billing process to ensure the coverage adjustments are commercially sound and comply with regulators’ demands.

The solution

The AI-powered cooperation for efficient networks Catalyst is designed to provide an OSS framework that automatically manages the necessary reconfiguration for CSPs involved in adjusting their coverage. It aims to develop Multicriteria Intelligent Resource Allocation (MIRA) to self-design future slice-based networks efficiently and on demand.

The solution uses AI to design resource allocation and deployment for telecommunication services and slices, which is a significant step towards fully autonomous, self-designing networks. It also helps the network provider to use resources efficiently and deliver quality services with minimized cost and time, while optimizing energy efficiency and network performance. In addition, MIRA allows different types of AI to work together towards a common goal, making it possible to redesign the network automatically to fulfil the intent defined by other AI deployments.

By enabling multiple CSPs to work together to agree on a single provider and eliminate redundant coverage, the project is reducing energy consumption and also helping to build momentum behind a carbon neutrality roadmap for industry and regulatory bodies.

The Catalyst makes use of a range of existing TM Forum-standardized production components, including Resource Catalog Management, Resource Order Management, Resource Inventory Management and Open APIs. For integration between components, the Catalyst uses Enterprise Integration, while employing the Resource Performance Management component to measure performance and energy consumption.

In its final phase, the Catalyst also seeks to integrate further components of the TM Forum Open Digital Architecture (ODA). The first of these will perform inter-CSP negotiations for splitting the responsibility for geographical coverage and determine the optimal coverage area between CSPs, while the other component will intelligently allocate network resources such as RAN and core network services which allow RAN elements to be switched off.

Applications and wider value

So far, the Catalyst has achieved impressive results, reducing operational costs for CSPs and the manual effort required to reconfigure the network. By dynamically adjusting network configuration to adapt to traffic conditions, CSPs have been able to easily modify and apply policies for network resource allocation based on benchmark values for factors such as cost efficiency, energy efficiency, energy greenness, and performance.

The Catalyst has also shown itself as a means to contribute to the industry-wide goal of net zero, effectively reducing energy consumption by ensuring that the minimum amount of network hardware is working during low-demand periods. According to Benoit Radier, Research Engineer at Orange, “our Catalyst helps network coverage go green through AI. We can now maximize CSP resource efficiency, and minimize energy consumption and negative environmental impact, thanks to this multi-stakeholder collaboration.”