Creating Value That Transforms the World: Access Networks to Support a Sustainable Society

1. Introduction

Looking back over the past 25+ years, we at NTT Access Network Service Systems Laboratories (AS Lab) have been devoted to research and development (R&D) in optical communication technologies, to drive the rollout of fiber-to-the-home (FTTH) services [1]. Optical communication, including FTTH, has dramatically transformed our lives, making them more convenient and comfortable alongside the widespread adoption of mobile phones [2]. Today, connectivity is taken for granted, and communication infrastructure has become an indispensable part of daily life.

In 2019, NTT announced its Innovative Optical and Wireless Network (IOWN) vision and has since been engaged in R&D to make it a reality. NTT aspires to lead in creating new value while contributing to a globally sustainable society [3]. With sixth-generation mobile communication systems (6G) on the horizon, AS Lab is united in accelerating R&D toward the implementation of IOWN and 6G.

One of the most significant environmental changes affecting us is the rapid advancement of artificial intelligence (AI). In an era of fast-evolving global trends and increasing uncertainty, AI is expected to play a key role in addressing challenges related to planetary sustainability [4]. AS Lab is actively pursuing “AI for NW”—the use of AI to enhance network operations—as well as “NW for AI”—research into how networks should evolve to make AI usage more efficient and effective. These two complementary research streams will continue to be a priority.

Domestically, the deterioration of infrastructure built during Japan’s period of rapid economic growth has become a pressing concern [5], and our communication infrastructure is no exception. With Japan facing both a shrinking and aging population [2] and an increasing frequency of natural disasters due to climate change [6], we are also accelerating R&D on smart infrastructure that will radically improve the efficiency of inspection, maintenance, and management of communication infrastructure. Looking ahead, we intend to apply the technical expertise we have accumulated in telecommunications to other types of infrastructure, contributing to the sustainability of the entire social infrastructure ecosystem. Our goal is to develop these capabilities into technologies that can compete globally.

In this context, AS Lab has defined the following mission, as shown in Fig. 1: “Taking on the challenge of creating new value through cutting-edge access network technologies and their practical application, thereby contributing to a globally sustainable society.” To achieve this, we are reinforcing efforts in robust networks, environmental impact reduction, and safety, while staying mindful of global perspectives and leveraging emerging technologies. Our R&D spans the five core access network domains—optical fiber access, infrastructure, access system, wireless access, and operations.

Fig. 1. AS Lab’s R&D strategy.

In the following sections, we introduce specific advanced technologies categorized under three strategic objectives: technologies supporting the advancement and diversification of services, technologies enabling fundamentally smarter network operations, and technologies for pioneering new business domains.

2. Technologies supporting the advancement and diversification of services

Figure 2 highlights technologies that contribute to the high-speed, high-capacity, and low-latency capabilities of optical and wireless systems, as well as those that enable more diverse service offerings.

Fig. 2. Technologies supporting the advancement and diversification of services.

The optical fiber technology for enhanced network performance shown in Fig. 2(a) supports ultra-high-capacity transmission by using multi-core optical fiber cables, which house multiple cores within a single fiber. This enables petabit-level optical transmission. As demand for optical fibers surges in both submarine and terrestrial networks, it is critical to sustainably expand transmission capacity within the constraints of existing facility space. This technology maintains compatibility by preserving the standard cladding diameter of 125 μm while achieving over four times the spatial utilization efficiency. Additionally, innovations in fiber structure design and manufacturing processes aim to expand the low-loss transmission range into the O-band, allowing for broader wavelength coverage. Collective amplification technology for multi-core optical fibers also contributes to environmental impact reduction.

The remote-control-enabled All-Photonics Network (APN) transceiver architecture illustrated in Fig. 2(b) supports on-demand, rapid service provisioning and efficient operations and maintenance. This is especially critical for low-latency inter-datacenter connections and the construction of wireless access networks in the Beyond 5G/6G era, where quick and flexible adaptation to changing user demands is essential. To meet these needs, we are researching and developing a method that does not rely on the format or protocol of the customer’s communication signal (main signal). Instead, it overlays a monitoring and control channel onto the same optical fiber as the main signal, allowing for remote path configuration, activation, and monitoring by a controller. We are also investigating implementation and power-saving techniques. This research has been supported by Japan’s National Institute of Information and Communications Technology (NICT) under the project JPJ012368G50201.

Figure 2(c) introduces the multi-radio proactive control technology Cradio^®, a technology for proactively sensing, predicting, and controlling wireless network conditions to ensure uninterrupted connectivity. As applications that cannot tolerate communication interruptions continue to expand, expectations for higher-quality wireless communications are growing. Cradio^® is being refined through field trials to meet real-world demands, addressing complex use cases that involve diverse and conflicting requirements such as high speed, large capacity, low latency, and coverage hole elimination. We are also expanding the scope of prediction, sensing, and control to support advanced applications such as autonomous driving, smart cities, and smart factories through the sophisticated combination of multiple radio access technologies. Looking ahead to the 6G era, we are exploring wireless sensing technologies for enhanced security, integration with cyberspace, and more adaptive wireless device control based on sensing data and environmental changes.

Figure 2(d) features a technology that enables high-speed antenna and beam search for millimeter-wave distributed multiple input multiple output (MIMO) systems, which are key to implementing vehicle-to-everything (V2X)^*1 communication. This technology ensures stable millimeter-wave communications even in high-speed and obstructed environments. For 6G-scale data transmission, distributed MIMO using high-frequency bands such as millimeter waves is considered promising. However, to deploy millimeter-wave distributed MIMO for high-speed moving objects such as vehicles, beam selection must be executed at high speed to track movement in real time. Traditional approaches stagger the beam search timing across antennas to avoid interference, but as the number of antennas increases, so does the beam search time. Our solution leverages the properties of orthogonal frequency-division multiplexing (OFDM)^*2 to suppress interference, enabling simultaneous beam search across all distributed antennas. This ensures stable communication under dynamic and obstructed conditions.

3. Technologies enabling fundamentally smarter network operations

Figure 3 illustrates technologies aimed at transforming operations and maintenance through advanced smartification.

Fig. 3. Technologies enabling fundamentally smarter network operations.

The remote optical node technology shown in Fig. 3(a) enables automated remote fault isolation, which eliminates the need to dispatch field technicians during network failures and contributes to faster recovery. In the event of sudden equipment or optical fiber failures or damage caused by natural disasters, it is critical to quickly identify the root cause and restore service. Identifying whether the issue lies in the optical fiber or equipment traditionally required dispatching personnel, resulting in time and labor costs. This new technology extracts part of the optical signal via a tap, enabling remote power measurement of the signal to isolate the fault location. By using image sensors to measure the power of multiple optical signals simultaneously and in real time, the overall measurement time can be significantly reduced.

The image-based infrastructure degradation prediction technology shown in Fig. 3(b) leverages AI not only to streamline and simplify current inspection processes but also to enhance maintenance quality. It helps reduce operational burden, lessen dependence on skilled labor, and standardize inspection quality. Social infrastructure faces growing challenges such as aging facilities, a shortage of skilled maintenance workers, and increasing maintenance costs. In response, we have developed an AI model capable of generating future corrosion progression images, enabling infrastructure degradation diagnostics and prediction using batch images captured from vehicle-mounted cameras or drones combined with AI analysis. This allows for optimization of inspection and repair timing and methods, resulting in substantial cost savings for maintenance.

The AI-powered network data correction and autonomous fault analysis technology in Fig. 3(c) aims to accelerate response times and enable zero-touch fault management in complex networks. Handling faults in large-scale networks requires analysis of massive datasets, but using AI services alone has often failed to produce accurate troubleshooting procedures. To address this, our approach breaks down historical response procedures into simpler sub-tasks, allowing generative AI to produce accurate response instructions. This enables faster and more autonomous fault handling. We also apply AI for detecting and correcting data inconsistencies across distributed datasets by leveraging information about network topology and relationships, improving mapping accuracy. This supports more effective use of AI in maintenance tasks based on integrated and high-precision network data.

4. Technologies for pioneering new business domains

Figure 4 showcases technologies that enable the creation of new business opportunities.

Fig. 4. Technologies for pioneering new business domains.

The optical-fiber-based environmental monitoring technology shown in Fig. 4(a) uses data analysis to quickly and accurately detect the early stages of road excavation work, leveraging existing optical fiber infrastructure. There has been a growing number of unauthorized road excavation activities, increasing the burden of monitoring operations and the risk of infrastructure damage. This technology analyzes environmental vibration data collected from existing optical fibers, comparing vibrations during normal conditions with those during construction. This allows for automatic detection of unreported excavation activity. The algorithm of this technology is highly resistant to environmental noise from vehicles or large facilities, minimizing false positives. Field trials are currently underway to validate this technology’s potential to reduce the burden of infrastructure monitoring and prevent damage caused by unauthorized construction.

Figure 4(b) presents four-dimensional (4D) infrastructure mapping technology that improves the accuracy of inspection and image capture locations and precisely calculates the coordinates of infrastructure facilities based on inspection imagery. When using dashcams (drive recorders) for inspections, the inherent inaccuracy of Global Positioning System (GPS)-based location data has made it necessary for personnel to manually match inspection images with the relevant infrastructure components, creating additional workload, especially problematic as the maintenance workforce shrinks. This technology automates the linkage between inspection images and infrastructure components, eliminating the need for manual mapping and improving inspection efficiency. By accumulating inspection data over time for each facility location, the technology also enables optimization of repair and replacement planning, contributing to reduced maintenance workloads across all social infrastructures.

5. Conclusion

This article outlined the direction of our R&D initiatives in access network technologies, along with key technologies under development, all aimed at creating new value and achieving global sustainability through cutting-edge innovations and practical implementations in access networks. We will continue to take on the challenge of advancing research and accelerating the practical deployment of IOWN, 6G, and related technologies.

References

[1]	History of NTT Access Network Service Systems Laboratories (in Japanese), https://www.rd.ntt/as/history/
[2]	Ministry of Internal Affairs and Communications, “2022 White Paper on Information and Communications in Japan,” https://www.soumu.go.jp/johotsusintokei/whitepaper/eng/WP2022/2022-index.html
[3]	NTT, Medium-Term Management Strategy “New Value Creation & Sustainability 2027 Powered by IOWN,” https://group.ntt/en/ir/mgt/managementstrategy/
[4]	Ministry of Internal Affairs and Communications, “2024 White Paper on Information and Communications in Japan,” https://www.soumu.go.jp/johotsusintokei/whitepaper/eng/WP2024/2024-index.html
[5]	Ministry of Land, Infrastructure, Transport and Tourism, “Infrastructure Maintenance Information” (in Japanese), https://www.mlit.go.jp/sogoseisaku/maintenance/02research/02_01.html
[6]	Cabinet Office Japan, “White Paper on Disaster Management,” https://www.bousai.go.jp/en/documentation/white_paper/index.html