The ‘End-to-end service realization using intent-based networks’ Catalyst demonstrates how TM Forum Open APIs and business intent can be employed to automate CSPs’ service management layer, including the rApps used to optimize network traffic. By moving beyond single use cases to automating the entire multi-layer network operation stack, the project aims to increase efficiency and reduce CSPs’ operational costs.
Scalable, reliable, efficient networks through intent-based management
Commercial context
CSPs have historically grappled with a difficult trade-off. In designing and running commercial networks, they can typically achieve only two out of the three key objectives: scale, reliability, and efficiency. They can never achieve all three at once. As the industry has evolved and customer expectations have increased, scale and reliability have become essential. As a result, this has left efficiency as the objective most often compromised. The resulting imbalance drives up operational costs and squeezes profit margins. Moreover, aggressive market competition makes the issue even more severe.
To overcome this persistent challenge, CSPs must rethink their entire approach to network lifecycle management. This means moving beyond task-driven automation toward intent-based autonomy, where business intents guide every phase of the network lifecycle. From planning and rollout to day-to-day operations and eventual decommissioning, decisions can be orchestrated through a unified knowledge layer that interprets and executes these strategic intents.
The goal of the ‘End-to-end service realization using intent-based networks’ Catalyst is to build such a self-adaptive system. This Catalyst implements TM Forum Open APIs and intent-based management across multiple layers to aid CSPs in achieving a transformation in the approach to automation. Ultimately, the Catalyst simultaneously achieves efficiency, enhanced user experience, and the Moonshot challenge of empowering Level 4 autonomous networks.
The solution
The approach to reaching autonomous network Level 4 involves integrating customer and operator intents within the service layer. Customer intent originates from product orders, specifying requirements for throughput. Operator intent originates from the independent introduction of energy considerations aligned with TM Forum's new soft requirements standards. The intents are translated into actionable requirements for rApps-based optimizers. Multiple rApps also coordinate via the non-real-time RIC.
The project demonstrates practical conflict resolution and mitigation mechanisms through real-world examples. Optimizers react to detected anomalies by adjusting configurations in a closed-loop fashion at the resource level. Service assurance monitors compliance with service intents and triggers healing actions to maintain performance standards. As a result, this closes the loop at the service level.
While the general Catalyst architecture can address various closed-loop and intent-driven scenarios, the Catalyst has a restricted focus. Specifically, it focuses on intent-based of optimization of network slices for video streaming during live events. Meanwhile, another intent is to execute a CSP strategy for energy efficiency. By translating customer and operator intents into automated network adjustments, the system effectively manages resources to support high-quality streaming. This ensures that video delivery meets performance requirements and efficient resource allocation.
The project’s main purpose is to demonstrate the feasibility of the latest TM Forum assets for product and service management in combination with intent management and the TM Forum Intent Ontology models. The project uses Open APIs TMF622, TMF921, TMF645, and TMF641. Operators use these APIs with the new capability to communicate and manage intent added in v5. Additionally, the project uses the latest TM Forum Intent Ontology, including the TR290x, TR291x and TR292x series of specifications. Furthermore, it follows the autonomous networks architecture in IG1251x and the intent definitions and concepts from IG1253 and IG1358, and the autonomous networks levels introduced in IG1252.
The project is also developing extensions to the intent models and intent APIs and contributing them back into the Autonomous Networks Collaboration project. As a result, it is expected to lead to a new specification for the expression of soft requirements in the TR291x series. Furthermore, the project is testing all used assets for gaps and inconsistencies and reporting the findings back to the ODA, Open API and Autonomous Networks Projects.
Wider application and value
The Catalyst demonstrates the ability to achieve a far greater level of automation and efficiency through intent-based network management. It converts customer and operator intents into automated network adjustments to reduce manual intervention and costs.It also introduces intent in rApps, allowing them to autonomously manage resources, self-optimize, and self-heal, minimizing human oversight and expenses. Furthermore, it coordinates actions between rApps, detaching and managing conflicts using real-world examples. Finally, it implements continuous monitoring and automatic configuration changes to maintain performance and cut operational disruptions.
The Catalyst demonstrates that the needs of diverse applications can be matched at consistent levels of quality through intent-based network management. Through intent, we can communicate the performance needs for these applications, while increasing the autonomy of systems and helping CSPs to achieve a self-healing, self-optimizing network. The automation will help reduce operational costs and enable new use cases for businesses, while intent can also be used to improve the energy efficiency of systems.
As Mark Sanders, Chief Architect and Executive Autonomous Networks Enablement, Telstra, stated, “this Catalyst demonstrates the practical implementation of TM Forum Open APIs and intent-based management across multiple layers - from business objectives through service management to rApp deployment. By spanning the entire network operations stack, it showcases how true network autonomy can be achieved beyond isolated use cases, enabling a comprehensive, business-value driven operational model.”