Fiber Technology for Submarine SDM
By Sergei Makovejs
January 19, 2020
Repeatered submarine optical fiber systems are unique in the sense that they have inherent limitations on how much electrical power can be remotely delivered from the shore to submarine cable repeaters. This is different to terrestrial fiber systems, in which there are no fundamental limitations on how much power can be brought locally to power transmission equipment, although the cost of power delivery can vary. Electrical power constraint is also the reason why submarine systems historically operated with a relatively low number of fibers, typically 12 fibers or less, consistent with the limit on the number of amplifiers that can be powered from the constrained source at the shore.
Due to such a limited number of fibers in the cable, the fiber has always been considered as a precious resource, and maximization of capacity per fiber has historically been a priority. As a result, submarine links have always relied on the latest innovations in systems and fibers such as, the lowest attenuation silica-core fibers with the largest possible effective area, and acceptable microbend and macrobend performance. To maximize capacity per fiber (referred to as the “traditional” approach throughout this article), the goal of submarine network engineer is to optimize launch power into the fiber to achieve highest signal-to-noise ratio (SNR) for any given transmission reach and data rate requirements. Figure 1 shows an example of what such an optimization process may look like for a transpacific 10,000 km link with 60 km repeater spacing for the two fiber types. Both fibers have the same typical 1550 nm attenuation of 0.150 dB/km, but different effective areas (Aeff) – 115 µm2 and 150 µm2 (also typical at 1550 nm). Due to its larger Aeff, 150 µm2 fiber allows for a 1 dB increase in optimum launch power relative to 115 µm2 fiber resulting in 1 dB SNR increase.
To continue reading the rest of this article, please read it in Issue 110 of the SubTel Forum magazine here on page 28.
