1. Cost-Optimized Submarine Cables Using Massive Spatial Parallelism
- Author
-
Herve Fevrier, Ronen Dar, Peter J. Winzer, Szilard Zsigmond, Stephen G. Grubb, K.-Y. Huang, and Andrew R. Chraplyvy
- Subjects
Optical amplifier ,Multi-core processor ,Computer science ,Amplifier ,Submarine ,02 engineering and technology ,Noise figure ,Multiplexing ,Atomic and Molecular Physics, and Optics ,020210 optoelectronics & photonics ,0202 electrical engineering, electronic engineering, information engineering ,Electronic engineering ,Electric power ,Massively parallel - Abstract
We study the techno-economics of submarine systems constrained by a fixed electrical power supply. We show significant cost savings for high-capacity submarine systems using massive space-division multiplexing (SDM), even without assuming any savings from SDM-specific subsystem integration. Systems with about 100 parallel optical paths, e.g., $\sim$ 50 fiber pairs are shown to provide minimum cost/bit, operating at reduced spectral efficiencies and deep within the linear regime. While advanced nonlinearity-optimized fibers and digital nonlinearity compensation schemes provide little to no gain in such systems, SDM integration of amplifiers and transponders is shown to be a source for significant additional cost savings. We further examine the permissible cost premium for multicore fibers in such massively parallel systems and revisit various design tradeoffs for optical amplifiers, showing that a reduced noise figure can be traded for better power conversion efficiency. We also evaluate potential gains from increasing the available electrical supply power and discuss reliability aspects of massively parallel submarine systems.
- Published
- 2018
- Full Text
- View/download PDF