As the nonlinear capacity limit of single mode fiber (SMF) transmission systems is being approached, space-division multiplexing (SDM) in multicore fibers (MCFs) or few-mode fibers (FMFs) is currently under intense investigations to achieve ultrahigh spectral efficiency per fiber. Meanwhile, a key advantage of SDM over simply increasing the number of SMFs, is its inherent device integration and resource sharing capability. This can potentially provide significant benefits in terms of the cost per bit in future optical networks. In order to efficiently address capacity scaling in a single optical fiber, few-mode and multicore erbium-doped fiber amplifiers are being developed. Critical for the implementation of SDM amplifiers is to achieve almost the same amount of gain for all spatial channels. In this respect, we have recently demonstrated multimode fiber amplifiers, supporting >15 modes, with a maximum differential modal gain of 2 dB and negligible mode mixing.
For the first time, we demonstrate the implementation of a core pumped few mode erbium amplifier utilizing a mode selective photonic lantern for spatial modal control of the pump light. This device is able to individually amplify the first six fiber modes with low differential modal gain. In addition, we obtained differential modal gain lower than 1 dB and signal gain of approximately 16.17 dB at λ<sub>s</sub> = 1550 nm through forward pumping the LP21 modes at λ<sub>p</sub> = 976 nm.
Characterization of spatial mode content and dispersion properties in fiber laser systems and space-division
multiplexing (SDM) applications is key to understanding fiber properties and system performance. Several
techniques exist for modal characterization but often present limitations in the context in which they can be
used efficiently. In this paper, we present a powerful analysis scheme that removes several of those limitations
and pushes modal content analysis to a new level.
Space-Division Multiplexing (SDM), introduced more than 3 decades ago, is currently subject to intense research due to
its ability to provide order-of-magnitude capacity growth in future transmission systems. Fibers play a central role in this
renewed research field, and significant efforts have recently been spent to develop new fibers for SDM. In this paper, we
will review the most recent advances on these different SDM fibers. We will also compare their performances and
evaluate their respective potentials using the spatial density parameter that is a measure of the space efficiency.
New generation systems are expected to include more intelligent amplifiers able to adapt to many conditions
including different gains, channel load, temperature, aging and transient events.<sup>1</sup> To face the challenge and
meet these new requirements, having an accurate control on the Er environment within the fiber core matrix
has never appeared to be so necessary and predominant as it is the case now. Unlike conventional solution
doping techniques where Erbium ions are randomly incorporated in the fiber core, our process makes use of
a soft chemical synthesis to initially produce Erbium-doped nanoparticles (NPs). Erbium ions are therefore
incorporated in the fiber core together with their local environment. So far, our investigations2 first showed that,
from the material point of view, quenching levels are intimately linked to the design of the NPs through their
chemical composition. Then, from the system perspective, we evidenced the higher power conversion efficiencies
exhibited by NP fibers when compared to their conventional counterparts in high power amplifier configurations.
In this paper, we address our most recent work focusing on the NP optimisation towards quenching-free Erbiumdoped
fibers with a particular focus on core-shell alumino-silicate NPs. Completing our first amplifier results
obtained in high power configurations, we also explore new NP fiber profiles that extend the range of their
applications. Gain and noise characteristics of typical WDM operating points serve as key indicators on the
benefits our NP doping process could provide.
In this paper, we present the realization and the characterization of high-bandwidth 80μm-core MMFs. Insertion loss,
modal bandwidth and its assessment by Differential Mode Delay (DMD) measurements and macro-bend-loss
measurements will be particularly detailed. System performances at 10Gbps and 20Gbps over 10s of meters are
investigated using the IEEE Spreadsheet model and a more complete physical.
In 2009, we introduced a new doping concept involving Al<sub>2</sub>O<sub>3</sub>/rare-earth nanoparticles (NP) in a MCVD-compatible
process finding potential applications in Erbium-, Ytterbium- or Erbium-Ytterbium-doped fiber
amplifiers and lasers.<sup>1</sup> This approach, motivated by the need for increased efficiencies and improved attributes,
is characterized by the ability to control the rare-earth ion environment independently from the core composition.
The NP matrix can therefore be viewed as an optimized sub-micronic amplifying medium for the embedded rareearth
ion. The first experimental evidence to support this idea is reported in a comparative study with a standard
process<sup>2</sup> where homogeneous up-conversion (HUC) and pair-induced quenching (PIQ) levels are extracted from
Er<sup>3+</sup> unsaturable absorption measurements. NP-based fibers are found to mitigate the effects of the Er<sup>3+</sup> concentration increase seen in standard heavily-doped fibers. This conclusion is particularly clear when focusing
on the HUC coefficient evolution since, for a given type of NP, its level is independent from the Er<sup>3+</sup> concentration
in the doped zone. In this paper, we address our most recent work completing these preliminary results. First,
we investigate the quenching signature of a new NP design and its behavior when incorporated in different core
matrices. The interplay is further analysed by relating this set of measurements to practical EDFA performances.
Gain and noise characteristics of typical WDM amplifiers operating points serve as key benchmarking indicators
to identify the benefits of NP Erbium-doped fibers in the wide variety of EDFAs implementations.
Designed to overcome the limitations in case of extreme bending conditions, Bend- and Ultra-Bend-Insensitive
Fibers (BIFs and UBIFs) appear as ideal solutions for use in FTTH networks and in components, pigtails or
patch-cords for ever demanding applications such as military or sensing. Recently, however, questions have been
raised concerning the Multi-Path-Interference (MPI) levels in these fibers. Indeed, they are potentially subject
to interferences between the fundamental mode and the higher-order mode that is also bend resistant. This
MPI is generated because of discrete discontinuities such as staples, bends and splices/connections that occur
on distance scales that become comparable to the laser coherent length. In this paper, we will demonstrate the
high MPI tolerance of all-solid single-trench-assisted BIFs and UBIFs. We will present the first comprehensive
study combining theoretical and experimental points of view to quantify the impact of fusion splices on coherent
MPI. To be complete, results for mechanical splices will also be reported. Finally, we will show how the single-trench-
assisted concept combined with the versatile PCVD process allows to tightly control the distributions
of fibers characteristics. Such controls are needed to massively produce BIFs and to meet the more stringent
specifications of the UBIFs.
Fibers used for high power delivery are designed to ensure single-mode operation (in order to guarantee good output beam quality), large effective areas (A<sub>eff</sub>) and resistance to bend-induced distortions (in order to avoid non-linear effects). For simple step index fibers, the maximum Aeff of the fundamental mode that can practically be achieved at 1.06μm is ~350μm<sup>2</sup>. All-solid-silica Bragg fibers with large cores were proposed as an alternative solution for high power delivery through their fundamental core mode. These fibers consist of a low-refractive index core surrounded by a multilayer cladding that acts as a Bragg mirror. The loss spectrum of such fibers consists of a concatenation of several transmission windows separated by high-loss peaks. Here, we simultaneously study, for the first time (at our knowledge), the bending impact on Bragg fibers for the three critical properties required for high power delivery: large A<sub>eff</sub>, single-mode propagation and low bend losses for the fundamental mode. Thanks to their specific guiding mechanism, A<sub>eff</sub> as large as ~500μm<sup>2</sup> at 1.06μm can be achieved in Bragg fibers, while maintaining single-mode operation and bend losses lower than 0.1dB/m. Our numerical results are validated by experimental measurements on a PCVD Bragg fiber with a 40μm diameter core.
After many years of expectations, Fiber To The Home (FTTH) has finally become a reality with a wide number
of projects already running worldwide and growing. Optical fiber is inevitably taking more and more importance
in our environment, but for many good reasons, the space we are truly willing or able to allocate to it remains
limited. These installation constrainsts have turned into additional requirements that need to be addressed for
both active and passive components.
If exceptional bending performances obtained without degrading backward compatibilities is a pre-requisite
to deployment success,1 other parameters also need to be carefully taken into account when designing the ideal
candidate for use in confined environments. Among them, one can cite the bend loss homogeneity over length
and bending directions, the resistance to high optical power under bending and the tolerance to modal noise.
In this paper, we present the design and performances of a bend insensitive fiber optimized towards more
space savings and miniaturization of components. In addition to exceptional bending performances - lower than
0.1 dB/turn over a 5 mm bending radius -, its design guarantees impressive homogeneity levels and enhanced
safety margins for high power applications while being still resistant to modal noise. Successfull cleave- and
splice-ability results are finally presented, making this fiber ideally suited for use in components, pigtails and
Experimental and theoretical investigations of self-pumped phase-conjugate (SPPC) resonators with four-wave mixing (FWM) in diode-pumped amplifiers are presented. A model that uses a transient treatment of FWM (in one-pass or two-pass geometries) allows us to analyze the temporal dynamics and the energy characteristics of these resonators in both injected and self-starting configurations. The influence of the input energy in the injected case, and the influence of the output coupler reflectivity and diode pump energy in the self-starting case are analyzed. In the self-starting case, the SPPC resonators demonstrate self-adaptive compensation of phase distortions and produce a TEM<SUB>00</SUB> mode in a single-longitudinal-mode pulse by dynamic gain-grating formation with a reasonable optical-optical efficiency.
Self-pumped phase-conjugate (SPPC) loop resonators using four-wave mixing (FWM) in solid-state gain media are investigated in both injected and self-starting configurations. A model, using a transient treatment of FWM with the appropriate boundary conditions imposed by the loop geometry, allows us to analyze the threshold condition for oscillation, the temporal dynamics and the energy characteristics of such resonators. The influence of the input energy in the injected configuration, and the influence of the different gain gratings and output coupler reflectivity in the self-starting configuration are also analyzed. A SPPC loop resonator using FWM in a flash-lamp- pumped Nd:YAG amplifier is experimentally investigated. In the injected case, a maximum phase-conjugate reflectivity of 42, and a maximum extraction of 47 mJ are obtained. In the self-starting case, a TEM<SUB>00</SUB> mode output of approximately 130 mJ in a 13 ns single-longitudinal-mode pulse is produced up to 30 Hz. Experimental and theoretical results are in good agreement.