(Image Credit: Norlando Pobre)
Broadband networks are investments intended to last for decades and must accommodate ever-growing traffic and higher bandwidth demands. Ensuring long-term reliability requires careful fibre and cable management, robust connections throughout the network, and forward-looking design choices. Operators must meet current requirements while planning for future upgrades and evolving service needs.
The optical fibre infrastructure deployed today supports voice, internet access and video streaming and will typically host multiple generations of transmission systems over its lifetime. Data traffic volumes will continue to rise sharply as the number of internet users and connected devices grows worldwide. Alongside higher device counts, average fixed broadband speeds are expected to increase significantly compared with previous years, driving the need for networks that can scale in capacity and performance.
New standards and technology
Advances in fibre-optic cabling and connector standards directly influence network planning and deployment. Modern single-mode fibres are engineered to operate across a broad wavelength range—from about 1260 nm to 1650 nm—to support a wider set of services. However, longer wavelengths are more sensitive to bending losses, which can degrade signal quality if fibres are subjected to tight bends. Although ITU-T G.652D fibres accept a minimum bend radius of roughly 20 mm, this is impractical for many customer premises installations.
To address this, ITU-T G.657 A2 bend-insensitive fibres were introduced for premises cabling. When stored or routed around a 10 mm bend radius, these fibres exhibit macro-bending loss that is 10 to 20 times lower than that of G.652D fibres under the same conditions. The adoption of bend-insensitive fibres in FTTH rollouts has led to more permissive PON recommendations, allowing operators to reduce some installation complexity and rely on less-specialized labor. While this can lower initial costs and speed deployments, it also raises the importance of careful material selection and thoughtful network architecture.
Proper fibre splicing and high-quality terminations require trained technicians. Skilled splicing personnel are costly and increasingly scarce, so network architectures that minimize the number of splice points—by concentrating splices and increasing splice density where possible—can lower operational costs. Such architectural choices must be incorporated during the initial network design to be effective.
Many FTTH business cases have historically emphasized rapid payback and minimized upfront expenditures. That focus sometimes resulted in relaxed specifications for optical cables and connectors and less rigorous installation practices. While reducing material and labor costs may seem beneficial short-term, these savings can become liabilities over the network’s lifetime as successive generations of transmission equipment place higher demands on the fibre plant.
Historically, brief outages for residential customers were often deprioritized, but customer expectations have evolved. Standards and technologies are responding by expanding the usable fibre spectrum and promoting higher-quality installations so that costly network investments remain viable for longer.
Building to meet future requirements
Next-generation PON technologies such as NG-PON2 under discussion at ITU-T will enable operators to expand FTTH bandwidth and reduce deployment cost by allowing more users or multiple operators to share the same fibre. NG-PON2 can be deployed as overlays on existing Gigabit PON (GPON) networks, but supporting these new services requires components and installations that can perform across a wider wavelength range.
NG-PON2 downstream channels are intended to operate in the 1600–1625 nm wavelength band. Current ITU-T and IEC performance standards for cables and connectors do not always account for these longer wavelengths, which means components specified only for shorter wavelengths may not deliver the required performance. For truly future-proof networks, operators should specify all components—cables, connectors and passive elements—for reliable operation up to 1625 nm. Standardization bodies are expected to address these needs in upcoming revisions of cable and connector standards.
In summary, operators must design and build networks with future requirements in mind because technology and service demands will continue to evolve. While the exact trajectory of change is uncertain, the likelihood of using wavelengths up to 1625 nm is high. Applying lessons from past deployments—investing in technician training, enforcing good cable management practices, and choosing connectors and components with the appropriate performance specifications—yields both immediate and long-term benefits for operators and subscribers.
Do you have additional recommendations for operators building future-proof networks? Share your thoughts in the comments.