Every blueprint ships with source code, sample data, and step-by-step instructions — ready to run on your Cloudera Platform
-
Blueprint Card
Overview
The Intent-Based RAN Energy Efficiency Blueprint provides a simulation-validated framework for evaluating AI-driven energy optimization strategies in 5G Radio Access Networks (RAN).
This blueprint integrates:
- VIAVI RAN Scenario Generator (AI RSG)
- VIAVI ADK simulation environment
- Cloudera AI-hosted Large Language Models via an OpenAI-compatible inference endpoint
- Closed-loop Planner and Validation agent architecture
The system simulates network behavior, generates energy-saving actions, validates those actions against QoS constraints, and applies safe optimizations iteratively.
This enables engineering teams to evaluate AI-assisted network control policies before deployment.
Business Value
Rapidly scale VIAVI and Cloudera AI across live networks, reducing energy costs without compromising service quality.
- Cost Savings - Lower network energy costs; Higher return on network infrastructure; Lower ongoing operational effort.
- Revenue Growth - Faster time to value.
- Cost Avoidance - Protect customer satisfaction and retention, Lower implementation risk, Scale without costly redesign
- Other - Support sustainability and compliance targets
Demo
This demo shows how Cloudera AI and VIAVI AI RSG use a closed-loop, dual-agent approach to reduce 5G RAN energy consumption while maintaining operator-defined QoS targets. A Planner agent recommends cell sleep/wake actions based on network KPIs and intent, while a Validator agent checks the simulated impact before approved changes are applied. Watch the demo
Key Features
- Intent-based optimization: operators specify QoS targets in natural language; the system translates intent into concrete network actions
- Dual-LLM closed loop: a Planner LLM proposes energy-saving actions, a Validator LLM reviews and approves each action before it is applied
- Simulation-backed safety: every candidate action is tested in the VIAVI AI RSG simulator — no action is applied without first passing simulation
- Model-agnostic: compatible with any OpenAI-compatible LLM endpoint (Cloudera AI Inference, NVIDIA NIM, OpenAI, and others)
- Private AI by design: LLM inference runs entirely on Cloudera AI, keeping sensitive network telemetry inside your environment
Use Case
Reducing RAN energy consumption while maintaining strict Quality of Service (QoS) guarantees is a critical engineering challenge. Aggressive energy-saving techniques, such as cell sleeping, can negatively impact throughput and user experience if applied incorrectly. This blueprint evaluates AI-generated energy optimization actions in a validated simulation loop to ensure:
- Energy efficiency improvements
- QoS preservation
- Safe and controlled optimization
Architecture
The system operates as a closed-loop optimization pipeline. Each iteration loads current network KPIs, generates candidate energy-saving actions via the Planner Agent, simulates their impact using VIAVI AI RSG, validates them through the Validation Agent, and applies the approved actions before advancing to the next simulation interval.
Your form submission has failed.
This may have been caused by one of the following:
- Your request timed out
- A plugin/browser extension blocked the submission. If you have an ad blocking plugin please disable it and close this message to reload the page.