Concepedia

Publication | Closed Access

Few-Mode Fibers for Mode-Division-Multiplexed Systems

299

Citations

25

References

2014

Year

Abstract

We describe the design trade-offs that are at stake when optimizing few-mode fibers (FMFs) that support a high number (≥ 6) of LP modes. We particularly detail the design of 6-LPmode fibers that allow to multiply the capacity by a tenfold factor (two modes being spatially non-degenerate and four modes being two times spatially degenerate). For low-differential-mode-groupdelay (low-DMGD) FMFs adapted to strongly-coupled modedivision-multiplexed systems, trench-assisted graded-index-core profiles can be optimized to have Max|DMGD| <;10 ps/km and undesired leaky LP modes appropriately cut off, while all guided LP modes show good robustness (Bend Losses <;10 dB/turn at 10 mm bend radius). Such low-DMGD FMFs being sensitive to process variability, we show how fiber concatenations can efficiently compensate for this issue and that values <;25 ps/km can realistically be reached. For weakly-coupled FMFs adapted to weakly-coupled mode-division-multiplexed systems, step-index-core profiles can be optimized to have large effective index differences, Δn <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">eff</sub> , between the LP modes (Min|Δn <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">eff</sub> | >1.0 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> ) to limit mode coupling and A <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">eff</sub> >~100 μm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> to limit intra-mode non-linearity with good mode robustness. For such weakly-coupled FMFs, sensitivity to process variability is small and main characteristics do not significantly change when variations are within the manufacturing tolerances. We also briefly discuss experimental validations.

References

YearCitations

Page 1