When we talk about sustainable telecoms infrastructure, the conversation often jumps straight to energy consumption, carbon emissions, renewable power or more efficient network equipment. These are all important, but the May 2026 issue of NTT Technical Review reminds us that sustainability also has a very physical dimension. It is about the towers, poles, ducts, maintenance holes, conduits, closures, cables, covers, coatings and materials that quietly support communications networks for decades.
This is a timely topic because many countries are facing the same infrastructure challenge. Assets built decades ago are ageing at the same time that maintenance budgets, skilled workers and inspection capacity are under pressure. The NTT articles mention major infrastructure incidents in Japan in 2025, including sewer pipe collapses and ruptured water conduits, as examples of what can happen when ageing infrastructure and limited maintenance resources collide. The point is not that these assets were neglected, but that even well-managed infrastructure can become difficult to sustain when deterioration accelerates and resources become constrained.
For telecoms, this matters because modern networks are only as reliable as the physical infrastructure that supports them. Fibre, mobile base stations, switching equipment and transmission systems all depend on civil infrastructure. Underground ducts, maintenance holes, steel towers, poles and bridge-mounted facilities may not be as exciting as 5G Advanced, AI-RAN or 6G, but they are essential to service continuity, resilience and safety.
NTT’s approach to sustainable infrastructure is interesting because it is not limited to one technology. It defines sustainable infrastructure around four requirements: safety, economy, resource recycling and worker satisfaction. Safety means avoiding accidents and ensuring that maintenance workers can operate safely, including in enclosed spaces or at height. Economy means keeping maintenance costs low enough for assets to remain viable over the long term. Resource recycling brings in the circular economy, including reuse and recycling of equipment and materials. Worker satisfaction recognises that even with robots, AI and automation, human workers will remain essential, so maintenance needs to be practical, efficient and less burdensome.
The framework presented by NTT divides R&D into four areas. The first is maintenance, or changing the present, by improving existing maintenance work and extending service life. The second is sensing, or knowing the present, by detecting the condition of infrastructure more efficiently and ideally remotely. The third is prediction, or knowing the future, by forecasting how infrastructure will deteriorate in different environments. The fourth is design, or changing the future, by using the knowledge gained from deterioration prediction to create longer-lasting, easier-to-maintain and more recyclable infrastructure.
This way of thinking is useful because it shows that sustainability is not just about replacing old assets with new ones. Renewal buys time, but the new infrastructure will also deteriorate eventually. The real challenge is to understand degradation mechanisms, detect deterioration early, repair at the right time, and design future infrastructure so that it needs less maintenance in the first place.
One of the more practical examples is NTT’s work on smart maintenance for steel towers. Steel towers are exposed to wind, rain, humidity, salt and other environmental factors, and rust can affect long-term structural integrity. Traditional rust removal using power tools, metal brushes or sandblasting can be labour-intensive, difficult in narrow spaces and challenging around bolts. NTT is investigating laser-based rust removal as a smaller, lighter and lower-recoil alternative that could also be combined with robotics and AI in the future.
The clever part is that the laser is not only being treated as a tool for removing rust. NTT is also studying how laser irradiation changes the steel surface itself. If the process can form a stable oxide layer and improve paint adhesion, it may help suppress rust recurrence and extend repair intervals. That would reduce both labour requirements and maintenance costs. The work combines practical surface preparation with deeper materials science, including first-principles calculations and machine-learning-based molecular dynamics to understand how iron oxides form during rapid heating and cooling.
Another important area is corrosion prediction inside maintenance holes. NTT owns around 680,000 communication maintenance holes in Japan, and these spaces house fittings that support communication cables. Maintenance holes can be humid, nearly sealed environments where rainwater or groundwater enters and stagnates. Depending on the water level, metal fittings may alternate between submerged and high-humidity conditions, creating complex corrosion behaviour.
The article on corrosion deterioration focuses on metal fittings inside maintenance holes, including the local corrosion that can occur where a communication cable is secured by string. Standard salt-spray and cyclic corrosion tests are useful, but they do not always reproduce the exact corrosion behaviour found inside a maintenance hole. NTT therefore studied an air/solution alternating test, which better simulates the repeated wet and air-exposed state inside these environments. The research showed that this test could reproduce local corrosion directly under the string-contact section, making it more relevant for understanding deterioration in real facilities.
This is where the move from periodic inspection to condition-based maintenance becomes important. If operators can predict which maintenance holes or components are at higher risk, they can inspect those earlier, while extending inspection intervals for lower-risk assets. That is a far better use of limited maintenance resources than treating every asset in the same way.
Plastic materials are another area that does not receive enough attention in telecoms infrastructure discussions. Plastics are used in cable sheathing, branch cable covers, closure housings and bundling materials. They are lightweight, easy to form, electrically insulating and corrosion resistant, but they can degrade outdoors due to light, heat, water and stress. Ultraviolet light can trigger photooxidation, heat can accelerate chemical reactions, water can leach out additives, and mechanical stress can help microcracks grow into larger cracks.
NTT’s work on accelerated ageing tests for plastics, using polypropylene as an example, is about reproducing real degradation mechanisms more quickly without creating unrealistic failure modes. This distinction is important. It is easy to make a test harsher, but a harsher test is not automatically a better test if the degradation mechanism no longer matches what happens outdoors.
The researchers are therefore looking at chemical and physical indicators, such as carbonyl index measured by FT-IR, oxidation induction time measured by methods such as chemiluminescence, and mechanical properties such as tensile strength and fracture strain. They are also looking at test cycles that combine UV, heat, water and stress, as well as warm-water immersion to accelerate additive leaching. This kind of work can help identify materials with better weather resistance and support infrastructure with longer service life.
The final article broadens the discussion from telecoms infrastructure to social infrastructure. NTT’s Civil Systems Project has long worked on cable tunnels, maintenance holes, conduits and bridge-mounted facilities. The historical evolution is notable: in the 1970s and 1980s, the focus was on product development and construction methods; after the Great Hanshin-Awaji Earthquake in 1995, seismic performance became a priority; now, with ageing assets, the focus has shifted towards efficient and sustainable maintenance.
A simple but effective example is the Tapered DIAmond Iron Cover for maintenance holes. Its surface pattern changes visually as it wears, allowing inspectors to judge wear more easily without measuring groove depth. Its design also improves abrasion resistance and extends the replacement cycle to around three times that of the previous design. This is a good reminder that innovation in infrastructure is not always about advanced AI or robotics. Sometimes, better physical design can make inspection easier, reduce lifecycle costs and extend asset life.
That said, AI does play an important role. NTT has developed image-based diagnostic technologies that can inspect, diagnose and predict deterioration. One example is technology that predicts the future progression of steel corrosion from images of infrastructure facilities, such as road bridges. By combining images with environmental data, the model can generate predicted images showing how corrosion may spread. In verification using telecommunications conduit facilities attached to road bridges, the technology predicted the increase in corrosion area several years ahead with an average error of less than 10%.
NTT is also applying telecoms infrastructure know-how to wider social infrastructure, including roads, bridges, tunnels, water and sewage systems. Using accumulated facility data, it has built AI models to estimate damage risk from disasters such as earthquakes, heavy rainfall and flooding. The article also highlights the use of synthetic aperture radar satellite data to detect early signs of underground cavities before surface damage becomes visible. This could allow wide-area screening of roads and help reduce the cost and labour associated with traditional ground-penetrating radar inspections.
There is an important lesson here for the telecoms industry. Network sustainability cannot be measured only at the level of watts per bit or carbon emissions from active equipment. Those metrics matter, but they do not capture the full lifecycle of infrastructure. A sustainable network also needs long-lived materials, efficient inspection, predictive maintenance, safer working methods, lower lifecycle cost, and better reuse and recycling.
This will become even more important as networks evolve. 5G, 5G Advanced and future 6G systems will require dense, distributed and resilient infrastructure. Edge computing, fibre densification, small cells, private networks, non-terrestrial connectivity and AI-native operations all depend on physical assets that must be deployed, protected, inspected and maintained. The more digital the network becomes, the more important the physical layer of infrastructure remains.
NTT’s May 2026 feature articles are therefore a useful reminder that sustainable infrastructure is not a single technology area. It sits at the intersection of materials science, sensing, AI, robotics, civil engineering, chemistry, laser technology, satellite monitoring and practical field operations. It also shows that telecoms infrastructure expertise can be valuable beyond telecoms, especially as wider social infrastructure faces similar ageing, resilience and maintenance challenges.
The future sustainable network will not just be more energy efficient. It will also be easier to inspect, safer to maintain, smarter at predicting deterioration, built from better materials, and designed with the full lifecycle in mind. That may not sound as glamorous as the latest radio interface or AI breakthrough, but without it, the networks we rely on every day cannot remain reliable, resilient or truly sustainable.
Related Posts:
- Free 6G Training: The Role of the All-Photonics Network (APN) in IOWN and Future 6G
- Telecoms Infrastructure Blog: NTT Technical Review Highlights ITU-T Work on Standards for Higher-Capacity Fiber
- Telecoms Infrastructure Blog: Powering Vodafone’s Mobile Network with Solar Energy in Germany
- Telecoms Infrastructure Blog: Deutsche Telekom's Hydrogen-Powered Mini-Masts
- Telecoms Infrastructure Blog: O2 Telefónica Deutschland's Sustainable Phone Mast in Bavaria
- Telecoms Infrastructure Blog: KDDI's Power Backup Solutions
- Telecoms Infrastructure Blog: Du's Solar Sites in UAE
