Page 22: of Marine Technology Magazine (November 2025)

SUBSEA FIBER NETWORKS AI using the latest generation of multi-degree recon? gurable op- in the subsea cable. tical add/drop multiplexers (ROADMs) supporting subsea- An alternative end-to-end optimization leverages the use of optimized features, such as ampli? ed spontaneous emission additional bands in the terrestrial backhaul ? ber spectrum, (ASE) spectrum power insertion into the subsea ? ber, spec- such as C+L or Super C+L, to enable more wavelength di- trum sharing, and optical channel monitoring, while also sup- vision multiplexing (WDM) channels (see ? gure 2). While porting the features needed for terrestrial networks. a subsea ? ber typically provides 4.5THz of WDM spectrum,

Optical pass-through at the CLS sites eliminates two-thirds C+L supports 9.8THz and Super C+L supports 11.6THz, or of the transponders needed for end-to-end connections, and 2.5x more than the subsea cable. reduces power and space usage at the CLS, where these are Combined with coherent wavelengths on the backhaul ? ber often at a premium. While the extra backhaul distance may operating at higher data rates than on the subsea ? ber due to marginally reduce the maximum data rate of each connection, shorter distances, leveraging C+L or Super C+L WDM can this is partially offset by the performance of the latest genera- reduce the number of backhaul ? bers needed by 50-75% com- tion of 140 GBaud coherent optics. pared to the number of subsea ? bers, providing signi? cant

Thus, a CLS-to-CLS connection at 800Gb/s per wavelength overall savings for leased backhaul capacity.

across the Atlantic will be able to achieve an end-to-end speed Operators can complement the end-to-end network co-opti- of 700Gb/s when connecting to the data centers at each end mizations described above by using a common transport solu- of the subsea-terrestrial link. This end-to-end optimization tion across their subsea-terrestrial networks. This can. in turn, provides important bene? ts to hyperscalers and AI companies deliver more than lower costs and the operational simplicity of that control both the subsea and terrestrial ? ber routes that common spares, training, and management; it also unlocks the connecting their data centers. bene? ts of end-to-end automation to plan, optimize, turn-up,

The growth in cloud-based and AI services also implies glob- and monitor the network.

al coverage and reach, balancing the fact that foundational AI Network automation can help reduce operational costs and models and training occurring in one region will be delivered enable faster turn-up of end-to-end subsea bandwidth, lever- via inferencing to other regions or continents. Overlay with aging zero-touch commissioning tools to automate subsea that with national data sovereignty requirements and regula- channel optimization and service con? guration, reducing the tions related to network ownership, and oftentimes, a hyper- manual time and effort needed to turn-up traf? c, while also scaler or AI service provider may need to partner with local ensuring optimal performance and resource usage. network operators to provide terrestrial backhaul connectivity The well-used expression “it’s a small world” takes on a new from a subsea cable to data centers in other countries. meaning in today, where AI applications will permeate the

This terrestrial backhaul connectivity from a CLS to a data globe. Ensuring cost-ef? cient and ubiquitous global connec- center is often provided by a Managed Optical Fiber Network tivity and providing equitable access to new AI-based applica- (MOFN) provided by a local CSP, which can sometimes come tions will help bring people, cultures, and societies together at high cost if extending subsea wavelengths across the terres- across continents; more than ever enabled by the fabric of trial network requires as many ? bers for the backhaul as exist end-to-end subsea terrestrial networks.

Figure 3

Reducing the cost of subsea traf? c backhaul by using more WDM spectrum in the terrestrial network using C+L.

Credit: Nokia 22 November/December 2025

MTR #8 (18-33).indd 22 MTR #8 (18-33).indd 22 11/20/2025 6:19:43 PM11/20/2025 6:19:43 PM