―Collaborates with NVIDIA, Keysight, and Samsung Research America to build a high-fidelity digital twin to power AI-driven optimizations and autonomous operations―
KDDI Corporation
KDDI Research, Inc.
On June 23, 2026, KDDI Corporation and KDDI Research, Inc. launched an initiative toward realizing a high-fidelity RAN Digital Twin, creating a virtual replica of real-world radio access networks. This foundational platform will enable safe and efficient AI-driven network optimization, autonomous operation, and AI-algorithm evaluation in the 6G era. To accelerate this vision, KDDI and KDDI Research, Inc., are collaborating with NVIDIA Corporation, Keysight Technologies, Inc., and Samsung Research America, Inc to jointly develop this environment.
Together, the five companies will advance this collaboration to build the RAN Digital Twin and propel AI-driven networks forward.
KDDI and KDDI Research have developed a technology in which multiple AI agents cooperate to perform area optimization at base stations, and are deploying this technology across commercial networks nationwide, working toward the practical realization of AI-driven autonomous networks. [
1] As the telecommunications industry looks toward 6G, the demands on network quality will intensify, making the continued advancement of AI-based area optimization essential.
Meanwhile, training modern AI requires massive datasets and continuous iterative experimentation. Performing this AI learning directly in live network environments introduces significant challenges in speed and safety. Through this collaboration, the companies will build a RAN Digital Twin that faithfully replicates real-world network conditions in a virtual environment. This enables AI to be trained and validated without impacting live networks, and allows large numbers of scenarios with varying conditions to be simulated in parallel.
The project will deliver a prototype to demonstrate the scalability of the RAN Digital Twin by the end of March 2028. Looking further ahead, by the end of FY2030, the companies aim to extend the RAN Digital Twin to support diverse use cases and validate its performance on KDDI's commercial network. KDDI and KDDI Research envision utilizing its AI factories as the computing engine to operate the RAN Digital Twin at scale, continuously enhancing communication quality and delivering an exceptional experience to customers.
■About the Joint initiative
The companies will build a RAN Digital Twin that faithfully reproduces the real-world environment in a virtual space, by combining (i) technology that simulates highly complex phenomena such as radio wave propagation and wireless network behavior with high fidelity, and (ii) technology that scales such simulation on general-purpose computing infrastructure.
The RAN Digital Twin is intended to be used primarily in the following two use cases.
Use Case 1: Training and evaluation platform for AI network optimization
Through this collaboration, we will build a "RAN Digital Twin" that faithfully replicates real-world network conditions in a virtual environment. This enables AI to be trained and validated without impacting live networks, and allows large numbers of scenarios with varying conditions to be simulated in parallel, dramatically increasing the number of trials. Furthermore, the platform enables proactive optimization by allowing "what-if" scenarios, such as future changes in the radio environment or traffic fluctuations.
Use Case 2: Field-trial platform for new AI-based RAN features
In the 6G era, a large number of sophisticated AI models are expected to be embedded into RAN equipment. The RAN Digital Twin will also serve as an evaluation platform that enables lab-based testing of new features and algorithms across diverse virtual environments before field deployment, contributing to quality assurance and significantly shortening development cycles. This platform will be used in particular to validate technologies known as the "AI air interface," in which AI predicts and assists various RAN processes to improve communication quality and reduce power consumption. This will enable customers to benefit from cutting-edge AI technologies more rapidly.
■Roles of Each Company
| Company | Role |
|---|---|
| KDDI | Providing commercial network data and driving commercial trials |
| KDDI Research |
See the blog below for more details: |
| NVIDIA | Providing high-fidelity digital twin infrastructure with NVIDIA Aerial Omniverse Digital Twin and NVIDIA accelerating computing platforms See the blog below for more details: |
| Keysight | Providing user equipment (UE) emulation technology to support the development and validation of the RAN digital twin See the blog below for more details: |
| Samsung Research America | Providing vRAN technology that realizes base stations through virtualization See the blogs below for more details: |
■Comments from Each Company
- Satoshi Konishi, President and CEO, KDDI Research, Inc.; Senior Director, KDDI Corporation
"In recent years, KDDI has been rated the world's No.1 in connected experience by a third party. [
2] As networks grow increasingly autonomous and AI-driven in the 6G era, we are determined to raise that quality even higher. With the RAN Digital Twin, we can train and validate AI not only against real-world scenarios but even against future ones yet to unfold, accelerating innovation and dramatically shortening time to market. We are delighted to collaborate with NVIDIA, Keysight, and Samsung Research America to build this foundation and deliver an even better communication experience to our customers."
- Soma Velayutham, VP of Telecoms and AI, NVIDIA Corporation
"6G will be born in simulation and our collaboration with KDDI, Samsung, and Keysight will pioneer this future. As networks become increasingly autonomous, more complex AI algorithms will be the critical engines driving them. To develop these algorithms safely and at massive scale, the NVIDIA Aerial Omniverse Digital Twin provides a high-fidelity "AI-RAN gym" where operators can rigorously train, simulate, and validate new models before ever deploying them to a live network."
- Kailash Narayanan, SVP and President of the Communications Solutions Group, Keysight Technologies, Inc.
"RAN Digital Twins are the essential gatekeepers for safe, AI-driven network automation, requiring a zero-risk validation environment before commercial deployment. Keysight combines its advanced digital twin and scale emulation tools with technology from KDDI, NVIDIA, and Samsung to simulate the precise conditions needed to validate complex AI algorithms. This collaborative framework gives operators the trusted foundation they need to innovate confidently, eliminating live-network risks and fast-tracking the transition to autonomous 6G."
- Charlie Zhang, EVP, Samsung Research America, Inc.
"As a global leader in today's commercial vRAN deployment, Samsung recognizes RAN Digital Twin technology as a vital tool to drive the future AI-native vRAN in the 6G era. We are thrilled to collaborate with KDDI, KDDI Research, NVIDIA, and Keysight on developing a high-fidelity RAN Digital Twin, with the goal of enabling safer and more efficient optimization for future AI-driven networks."
- [1]
- [2]"Connected Experience" No.1 worldwide is based on Opensignal's Global Mobile Network Experience Report, in which it ranked first in the "Reliability Experience" category.
It achieved consecutive No.1 rankings in the two most recent reports (February 2025: comparison of 142 MNOs across 41 countries; February 2026: comparison of 151 MNOs across 44 countries).
For more details, please refer to the Opensignal website.
- *The information contained in the articles is current at the time of publication.
Products, service fees, service content and specifications, contact information, and other details are subject to change without notice.
Downloads