How Ericsson is Upgrading NetCloud With AI For Enterprises
Enterprise network management is changing as AI technologies mature.
Network administrators face mounting complexity from hybrid cloud environments, remote work infrastructure and increasing security threats, while IT budgets remain constrained. This has created demand for tools that can automate routine tasks and provide faster problem resolution.
In response to these market pressures, telecommunications equipment manufacturers are integrating AI capabilities into their network management platforms.
These systems aim to reduce the manual effort required for network configuration, monitoring and troubleshooting tasks that traditionally consume significant IT resources.
Against this backdrop, Ericsson Enterprise Wireless Solutions, has integrated a LLM into its NetCloud management platform to help network administrators automate configuration and troubleshooting tasks.
NetCloud security focus drives adoption of internal LLM technology
Unlike typical chatbot implementations that connect to external services like OpenAI's ChatGPT through application programming interfaces (APIs), Ericsson hosts its language models within its own infrastructure.
This approach prevents sensitive network data from being exposed to third-party services.
The system analyses data from the customer's network environment alongside Ericsson's technical documentation to generate responses tailored to specific network configurations.
This capability also aims to reduce the time network administrators spend correlating information from multiple sources.
- Ericsson has integrated LLMs into its NetCloud platform for enterprise network management, hosting the models internally rather than using external APIs like ChatGPT
- The system, called ANA (AI-based NetCloud Assistant), combines network data with technical documentation to automate troubleshooting and configuration tasks
- The platform uses an "agentic architecture" where multiple AI agents work in parallel to solve complex network problems
- A new feature will translate plain English business requirements into technical network policies automatically
Network administrators can now receive configuration instructions based on established practices rather than manually reviewing technical documentation when deploying new wide area network edge devices.
The system additionally assists with troubleshooting connectivity issues across software-defined wide area networks (SD-WAN) and security gateways, eliminating the need for manual port scanning and speed testing procedures.
Ericsson NetCloud expands AI operations features
The platform now includes an operations dashboard that establishes baseline measurements for network traffic patterns.
The system monitors variations in network performance metrics such as latency and jitter, which measure the delay and consistency of data transmission.
When the system detects anomalies, it categorises them by severity and identifies affected sites, users and applications.
The dashboard then provides analysis of the root cause, such as application server performance issues or network congestion.
"These insights in cellular network performance and application quality of experience empowers IT teams to proactively address issues to ensure optimal network operations," says Ericsson in a blog.
Ericsson's AI-powered network management system
The company has also begun demonstrating a feature that translates business requirements into network policies.
The system interprets plain language descriptions of business needs and generates corresponding traffic management rules.
These capabilities form part of what Ericsson terms an "agentic architecture", where multiple AI systems work in parallel to solve complex network management tasks.
The architecture enables virtual experts to execute and verify multiple operations simultaneously.
The network operations dashboard is also available to customers using NetCloud SASE (Secure Access Service Edge) and NetCloud Exchange services with appropriate licensing. Ericsson plans to release a version for wireless WAN customers in the first half of 2025.
The dashboard currently identifies latency and jitter anomalies specific to each customer's network, with packet loss detection planned for future releases.
For each detected issue, the system calculates the potential impact across the network infrastructure and suggests remediation steps.
āFirst of all, I believe AI is part of an evolution. AI will contribute in the future and we are seeing that contribution starting already today.
"What Gen AI will bring is an evolution of security toward the future, where you can actually detect new things with AIā, Keijo Mononen, Head of Security Solutions at Ericsson said last year.
Explore the latest edition of Technology Magazine and be part of the conversation at our global conference series, Tech & AI LIVE.
Discover all our upcoming events and secure your tickets today.
Technology Magazine is a BizClik brand
