In NTT at MWC26: Networks That Can Think, we looked at how NTT is working with partners to develop digital networks that do more than transport data. We explored how In‑Network Computing, or INC, doesn't treat communication and processing as separate stages; rather, the network itself helps to decide how data is handled.

The Way We Use AI

A key driver behind INC is the recent change in the way we use artificial intelligence. AI systems are no longer just used in data centers; now, they enable cameras to monitor infrastructure, robots to navigate warehouses, immersive systems to respond to human movement, and industrial platforms to watch for faults in real time.

In real time is the key here: when raw data has to travel long distances before decisions can be made, systems become harder to scale and more expensive to run. Making the network part of the processing loop helps to reduce those constraints and use computing resources more effectively.

Mobile World Congress

NTT will be showcasing its smart network research, developed in collaboration with the University of Tokyo and NEC, at this year’s Mobile World Congress in Barcelona.

Alongside that work, visitors will also be able to see a more specific demonstration: a low‑latency AI video analysis project jointly run by NTT DOCOMO and NTT. Positioned as a building block for 6G, it connects a commercial 5G standalone core to the Innovative Optical and Wireless Network (IOWN) All-Photonics Network (APN), integrating geographically distributed GPU resources under network-level control.

AI Inference

Inference is when a trained AI model looks at new data and makes a decision, such as identifying an object in a video. Until now, inference has typically been controlled outside the network—data travels across the mobile infrastructure, reaches a data center, and is processed there. When inference is spread across multiple GPUs, communication delays between them can affect performance, which is why high‑performance workloads are often kept in the same physical location. Doing so, however, can limit flexibility and increase infrastructure costs.

The INC Edge

DOCOMO and NTT have developed a function called the "INC Edge" within the 5G core. It connects the mobile network to the IOWN APN and coordinates AI inference processing alongside traffic control. Rather than simply forwarding packets, it helps determine how and where inference tasks are executed, allowing distributed GPU resources to be managed as part of the mobile network itself.

In a recent practical demonstration, video captured by a device was transmitted over commercial 5G. The INC Edge then directed that data through the APN to remote GPU servers, where AI video analysis was carried out. Priority control functions within the 5G core were applied to AI‑related traffic, while the APN provided wide bandwidth and low latency connectivity.

Another innovation was the division of inference into stages. Pre‑processing was handled before forwarding data onward, and only the necessary intermediate data was sent across the APN to remote GPUs. This reduces unnecessary transfer and allows geographically separated computing resources to be used more efficiently.

In applications that produce large volumes of video or sensor data and need quick analysis, coordinating distributed computing resources while maintaining quality allows operators much more freedom in where they place GPUs and how they manage capacity.

The Path to 6G

It also fits nicely within current thinking about 6G. Future mobile networks are expected to integrate communication and distributed computing much more tightly than today’s systems; by coordinating inference within the mobile core and across the APN, DOCOMO and NTT's project shows how that integration might work in practice.

You’ll be able to see all of this for yourself at Mobile World Congress Barcelona 2026, which runs from March 2 to March 5. NTT, NTT DOCOMO, and NTT DATA will be presenting together at Fira Gran Via – Hall 3, Stand 3M29, showcasing how photonics, AI, and next-generation networks are beginning to converge in practical ways. If you’re planning to pay MWC a visit, stop by to see the system in action and talk to the engineers behind it. ¡Pásense a saludar, amigos!

Innovating a Sustainable Future for People and Planet

MWC 2026 Official Site:
https://www.mwcbarcelona.com/

NTT Group at MWC
https://www.global.ntt/mwc/

If you have any questions on the content of this article, please contact:
Public Relations
NTT Information Network Laboratory Group
https://tools.group.ntt/en/news/contact/index.php

Daniel O'Connor

Daniel O'Connor joined the NTT Group in 1999 when he began work as the Public Relations Manager of NTT Europe. While in London, he liaised with the local press, created the company's intranet site, wrote technical copy for industry magazines and managed exhibition stands from initial design to finished displays.

Later seconded to the headquarters of NTT Communications in Tokyo, he contributed to the company's first-ever winning of global telecoms awards and the digitalisation of internal company information exchange.

Since 2015 Daniel has created content for the Group's Global Leadership Institute, the One NTT Network and is currently working with NTT R&D teams to grow public understanding of the cutting-edge research undertaken by the NTT Group.