By Aether SMaRT-5G Team
Aether SMaRT-5G team successfully demonstrated proof-of-concept of a new energy saving solution with technology partners in Japan
As 5G networks grow, so do the challenges and the costs. Mobile operators are under pressure to not only deliver faster and more reliable connectivity but also to do it more efficiently. With energy costs rising and sustainability on everyone’s agenda, finding smarter, leaner ways to manage networks has become a priority.
At Aether SMaRT-5G, we’ve been working to meet that need. And today, we’re excited to share the results of a successful collaboration that proves you don’t need to wait for the full O-RAN stack to be operational to start reaping the benefits of intelligent automation.
Together with our technology partners in Japan, we’re excited to share the results of our proof-of-concept open sourced solution that demonstrates how traffic steering and energy savings can be achieved without using E2 or A1 interfaces, which are the key pillars of the O-RAN architecture.
Let’s break down what this means, why it matters, and what comes next.
The Problem: Complexity Slows Innovation
Open RAN architecture promises a flexible and automated network built on open interfaces and AI-powered applications. But the typical O-RAN architecture relies on multiple components such as Service Management and Orchestration (SMO), Non-Real-Time (Non-RT) RIC, Near-Real-Time (Near-RT) RIC, and associated open interfaces.
For operators still rolling out their O-RAN based infrastructure, this can be a heavy lift. Many, like Rakuten Mobile, are attempting to achieve the benefits of O-RAN, starting with Non-RT RIC and SMO while the interfaces around the Near-RT RIC are getting matured to be deployable. Without a Near-RT RIC, deploying advanced energy optimization with traffic steering has been difficult, until now. The performance cannot match the one provided by a Near-RT RIC and xApps, but the complexity tradeoff might be worth it.
The Breakthrough: Smarts Without the Stack
What if mobile network operators could start optimizing energy consumption with traffic management today, without waiting to deploy the full O-RAN stack?
That’s exactly what we set out to prove. And we did.
In a proof-of-concept demonstration with technology partners Tietoevry, Rimedo Labs, Intel Labs and VIAVI Solutions at Rakuten Mobile’s Yokosuka Telecom Research Park RIC research facility in Japan, we deployed two rApps, Traffic Steering (TS) (details here) and Energy Savings (ES), working entirely through the O1 and Open Fronthaul M-Plane interfaces.
No E2. No A1. No Near-RT RIC.
The Results Speak for Themselves:
- +9% network throughput improvement via smart load balancing
- -9% energy consumption, with less than 1% impact on traffic
- Operation of both rApps, without needing A1
- Integration with open source SMO and Non-RT RIC platforms using only O1 interface
Why This Matters: Practical Automation, Now
This approach opens the door to real-world automation for operators at any stage of their O-RAN journey. Instead of waiting for all the pieces to be in place, they can start optimizing performance and energy consumption jointly today using lightweight O-RAN deployments.
This means:
- Lower deployment complexity
- Faster time to value
- Better sustainability outcomes
- A smoother path to full O-RAN architecture in the future
The Team Behind the Innovation
This breakthrough milestone was only possible thanks to a strong collaboration across an ecosystem that worked together to realize the vision:
Rakuten Mobile & Rakuten Symphony – Provided lab infrastructure and network requirements
Tietoevry – System Integration
Rimedo Labs – Developed the Traffic Steering rApp
Intel Labs – Developed the Energy Savings rApp
VIAVI Solutions – Provided commercial-grade RAN simulation and modeling
Aether SMaRT-5G / Linux Foundation – Led architecture, integration, and project oversight via open source platforms
Open Networking Foundation (ONF) – Utilizing the close tie between Aether SMaRT-5G and ONF, this project opened a pathway to field test and operationalize some of the results of the R&D project ONF has been carrying out in collaboration with Rutgers University, WINLAB.
The project has been focusing on innovative measurements and modeling of energy consumption in O-RAN, which has been made possible through a $2 million grant from the US government’s Public Wireless Supply Chain Innovation Fund (PWSCIF), administered by the National Telecommunications and Information Administration (NTIA).
What’s Next
We’re not stopping here. Deployment staging and phasing is already underway, and it will focus on production readiness for mobile operators, including:
- Further deployment testing in mobile operator SMO and Non-RT RIC environments
- Integration with real RAN hardware
- Enhanced AI/ML algorithms using insights from the NTIA PWSCIF-funded R&D project
- Full lifecycle testing in production-like environments
Our goal is to make this solution production-grade, scalable, and even smarter, delivering energy savings and performance gains at scale.
Open Source and Open Access
The best part? The work is open source and ready to explore. While the internals of the rApp algorithms remain proprietary, the full deployment framework, complete with Kubernetes and Docker configurations, simulators, and integration logic, is available in the project repository.
You can read our original announcement here or dive straight into the technical documentation to see how it works.
Final Thoughts: Less Stack, More Impact
This project proves that you don’t need the full O-RAN stack to make an impact. By working with a lightweight O-RAN deployment, we can accelerate the path to smarter, greener, open and more efficient networks.
If you’re an operator looking to optimize without overhauling, or a developer looking to contribute to meaningful, standards-based innovation, now is the time to get involved.
Want to connect or learn more?
Learn more at aetherproject.org reach out to us at info@aetherproject.org
