Description
The Autonomous Network Level (ANL) framework, standardized by 3GPP, provides a structured methodology to assess and communicate the degree of automation within a telecommunications network. It is not a specific technology or protocol, but a maturity model defined across multiple dimensions, including service deployment, assurance, and lifecycle management. The core of ANL is a six-level scale, where Level 0 (L0) represents fully manual, human-driven operations, and Level 5 (L5) represents a fully autonomous network capable of self-configuration, self-healing, self-optimization, and self-evolution with minimal to no human intervention. Each level is characterized by specific capabilities and the extent to which human operators are involved in decision-making and execution loops.
Architecturally, ANL assessment is integrated into network management systems, such as the Management Data Analytics Service (MDAS) and the Network Data Analytics Function (NWDAF) in 5G. The framework evaluates key automation domains: service deployment and provisioning, quality of service and experience assurance, energy efficiency, and security management. For each domain, a set of evaluation criteria and key performance indicators (KPIs) are defined in specifications like 3GPP TS 28.100 and TR 28.910. These criteria measure the network's ability to perceive its environment, analyze data, make decisions, and execute actions autonomously.
How it works involves a continuous cycle of measurement, analysis, and benchmarking. Network management functions collect operational data and telemetry. Analytics engines then process this data against the predefined ANL criteria for each domain. The system evaluates the degree of automation achieved—for instance, whether fault detection is manual (L1), automated with human approval (L3), or fully automated with proactive prediction (L5). The result is a composite ANL score or a profile showing levels per domain, giving operators a clear, standardized view of their automation maturity. This enables targeted investments in AI/ML and orchestration platforms to progress to higher levels.
ANL's role is pivotal for transforming network operations from reactive, human-centric models to proactive, software-driven paradigms. It provides a common language for vendors and operators to align on automation roadmaps and capabilities. By quantifying autonomy, it helps in validating the return on investment for automation technologies and ensures that different network slices or services can be operated with defined levels of autonomy, supporting the broader goals of zero-touch network and service management (ZSM).
Purpose & Motivation
ANL was created to address the growing complexity and operational cost challenges in modern 5G and future 6G networks. As networks become more software-defined, virtualized, and sliced, traditional manual management becomes impractical, error-prone, and slow. The industry needed a standardized, objective way to measure automation progress, moving beyond vendor-specific claims. ANL provides this benchmark, enabling operators to plan their evolution from legacy OSS/BSS to AI-driven autonomous networks systematically.
Historically, network automation was implemented in isolated silos without a unified maturity scale, making it difficult to compare solutions or set industry-wide goals. The limitations of previous approaches included a lack of clear metrics to gauge the effectiveness of automation investments and an inability to orchestrate cross-domain autonomous behaviors. ANL solves these problems by establishing a multi-dimensional framework that covers the full lifecycle of network services, from planning to optimization and healing.
The motivation for ANL's introduction in 3GPP Release-17 stemmed from the urgent need to support zero-touch network operations, reduce OPEX, and enable new agile services like network slicing. It aligns with broader initiatives such as ETSI's Zero-touch network and Service Management (ZSM) and the TM Forum's Autonomous Network framework, providing the 3GPP-specific technical definitions and integration points needed for mobile networks. By defining what each level means in practical terms, ANL helps the industry transition towards self-driving networks capable of meeting stringent SLA requirements for diverse use cases.
Key Features
- Defines a six-level maturity scale (L0 to L5) for network autonomy
- Provides multi-domain evaluation covering service, resource, and energy management
- Standardizes key performance indicators (KPIs) and evaluation criteria for automation
- Enables benchmarking and roadmap planning for operator automation journeys
- Integrates with 3GPP management architectures like MDAS and NWDAF for data collection
- Supports the vision of zero-touch network and service management (ZSM)
Evolution Across Releases
Introduced the initial ANL framework with definitions for the six autonomy levels (L0-L5) and high-level requirements. Specified the evaluation dimensions and key characteristics for each level in TS 28.100 and TR 28.910, establishing the foundation for measuring network automation maturity in management and orchestration.
Defining Specifications
| Specification | Title |
|---|---|
| TS 28.100 | 3GPP TS 28.100 |
| TS 28.910 | 3GPP TS 28.910 |