We experimentally investigate the performance of WDM coherent polarization-division multiplexed-return to zero-quadrature
phase shift keying (PDM-RZ-QPSK) network in nonlinear transport regimes. Seven, 28 or 32-Gbaud PDM-RZ-
QPSK channels are employed on a 50-GHz grid and transmitted over 1600-km fiber on an all-EDFA recirculating
loop without any dispersion compensation module (DCM). The transmission link is configured entirely of either
standard single-mode fiber (AllWave), medium dispersion fiber (TrueWave REACH), or ultra-large area fiber (ULAF).
We sweep the launch power of the center channel and side channels together to measure the nonlinear effects of self-phase
modulation (SPM), cross-phase modulation (XPM), and cross-polarization modulation (XPolM) on the center
channel's BER performance. Furthermore, for all link configurations, we employ three different carrier phase recovery
methods in the demodulation routine - Viterbi-Viterbi, Viterbi-Viterbi with a minimum mean-squared error (MMSE)
filter, and the Optametra/Tektronix Weiner filter - to ascertain their relative performance in the presence of nonlinear
Nonlinearities are a performance limitation in coherent optical links, and efforts have been made to understand the
tradeoffs between launch power and the penalties related to nonlinearities. Using both simulation and experimental
results from our 100G testbed we investigate the use of a nonlinear phase criterion that quantifies the total nonlinear
phase accumulation within a 112 Gb/s PDM-QPSK link. We examine the nonlinear effects of self-phase (SPM) and
cross-phase modulation (XPM) on a 112 Gb/s PM-QPSK channel propagating between four 10 Gb/s OOK aggressor
channels on a 50 GHz grid and quantify the launch power and span count scaling behavior. In order to assess the
applicability of a nonlinear phase criterion on real-world links, we determine the launch power that yields a 1.5 dB
OSNR penalty at a BER of 10<sup>-3</sup> for each configuration. This launch power then allows the identification of a Nonlinear
Threshold Power (number of spans times launch power) that fully incorporates the increasing nonlinear penalties with
further transmission distance. This metric allows for the determination of a set of engineering rules for deployment of
100 Gb/s PDM-QPSK in linear links with arbitrary number of spans and span distances. We find that this nonlinear
threshold is constant in dispersion-compensated links. These experimental results are validated with simulations.
Reliable simulations of high-speed fiber optic links are necessary to understand, design, and deploy fiber networks.
Laboratory experiments cannot explore all possible component variations and fiber environments that are found in
today's deployed systems. Simulations typically depict relative penalties compared to a reference link. However,
absolute performance metrics are required to assess actual deployment configurations. Here we detail the efforts within
the Georgia Tech 100G Consortium towards achieving high absolute accuracy between simulation and experimental
performance with a goal of ±0.25 dB for back-to-back configuration, and ±0.5 dB for transmission over multiple spans
with different dispersion maps. We measure all possible component parameters including fiber length, loss, and
dispersion for use in simulation. We also validate experimental methods of performance evaluation including OSNR
assessment and DSP-based demodulation. We investigate a wide range of parameters including modulator chirp,
polarization state, polarization dependent loss, transmit spectrum, laser linewidth, and fiber nonlinearity. We evaluate 56
Gb/s (single-polarization) and 112 Gb/s (dual-polarization) DQPSK and coherent QPSK within a 50 GHz DWDM
environment with 10 Gb/s OOK adjacent channels for worst-case XPM effects. We demonstrate good simulation
accuracy within linear and some nonlinear regimes for a wide range of OSNR in both back-to-back configuration and up
to eight spans, over a range of launch powers. This allows us to explore a wide range of environments not available in
the lab, including different fiber types, ROADM passbands, and levels of crosstalk. Continued exploration is required to
validate robustness over various demodulation algorithms.
This paper investigates DQPSK transport using both simulation and experimental results from our 100G testbed. We
examine 56 Gb/s single polarization (single-pol) RZ-DQPSK and 112 Gb/s polarization multiplexing (POL-MUX) RZDQPSK
with 12 Gb/s OOK aggressor channels and a variety of dispersion management maps using AllWave® zero
water peak (ZWP) fiber. Although a number of studies of 40 Gb/s line rates within 10 Gb/s networks have been reported,
there has been little with respect to 28 Gbaud DQPSK formats. We quantify the OSNR penalty due to nonlinearities of
these hybrid optical links. Using a nominal span loss of 22 dB and different span lengths while keeping the dispersion
compensation per span constant and the loss per span constant allows a direct examination of the impact of the residual
dispersion per span (RDPS) on the nonlinear penalty in the DQPSK channel. We vary compensation from 90% - 110%
(of total dispersion) across 8 spans (-119 ps/nm - +153 ps/nm). We report the required OSNR to achieve a non-FEC
BER of 10<sup>-4</sup> versus RDPS for both single- and dual-polarization (dual-pol) RZ-DQPSK. Experimental data is validated
against RSoft OptSim simulations.
There is an increasing demand for DWDM systems to support a set of network requirements (span length, total distance, capacity, etc) typically associated with previously distinct network application categories such as access, metro, regional, and long haul. This has led to the development of a multi-haul DWDM system, defined as an agile optical platform, able to combine multiple versions of transponders, amplifiers, and optical add/drop multiplexers (OADM) on the same optical link. The concept of a multi-haul system is applied in a study of DWDM networks that can employ, besides fixed OADM, two types of reconfigurable OADM (ROADM): a two-degree ROADM based on arrayed-waveguide gratings (AWG) and a multi-degree ROADM built with 5x1 wavelength-selective switches (WSS). A cost analysis on a network example calculates the savings obtained if an optimum mix of ROADMs is used on the same DWDM ring instead of deploying one single OADM type at all nodes. A technical feasibility study is performed to validate the performance of a DWDM system with up to 16 ROADMs in a network. To this purpose, OSNR penalties are experimentally determined for AWG and WSS ROADMs due to cascaded bandwidth narrowing, and crosstalk effects, with wavelength detuning values corresponding to locked and unlocked lasers.
A new fiber optic sensor integrating dielectric diffraction gratings and thin films on optical fiber endfaces is prosed for biomedical sensing applications. This device utilizes a resonant dielectric waveguide grating structure fabricated on an optical fiber endface to probe reactions occurring in a sensing layer deposited on its surface. The operation of this sensor is based upon a fundamental resonance effect that occurs in waveguide gratings. An incident broad- spectrum signal is guided within an optical fiber and is filtered to reflect or transmit a desired spectral band by the diffractive thin film structure on its endface. Slight changes in one or more parameters of the waveguide grating, such as refractive index or thickness, can result in a responsive shift of the reflected or transmitted spectral peak that can be detected with spectroscopic instruments. This new sensor concept combines improved sensitivity and accuracy with attractive features found separately in currently available fiber optic sensors, such as large dynamic range, small sensing proximity, real time operation, and remote sensing. Diffractive elements of this type consisting of a photoresist grating on a Si<SUB>3</SUB>N<SUB>4</SUB> waveguide have been fabricated on multimode optical fiber endfaces with 100 micrometers cores. Preliminary experimental tests using a tunable Ti:sapphire laser indicate notches of 18 percent in the transmission spectrum of the fiber endface guided-mode resonance devices. A theoretical analysis of the device performance capabilities is presented and applied to evaluate the feasibility and potential advantages of this bioprobe.
High-efficiency resonance coupling effects in zero-order diffractive multilayer structures have applications in fields such as optical filtering and laser technology. These resonance effects arise on phase matching of an incident laser beam to a leaky waveguide mode. Then, in theory, complete energy exchange between the input wave and a reflected wave can take place within narrow ranges in wavelength, angle of incidence, index of refraction, or layer thickness. This paper addresses theoretical modeling, experimental realization, and applications of this so-called guided-mode resonance (GMR) effect. In particular, the achievable GMR-filter efficiencies, spectral linewidths, sideband levels, and polarization characteristics are treated with a plane-wave model and a Gaussian-beam model. Resonance bandpass filters operating in reflection and transmission are shown to exhibit high efficiencies and extended low sidebands. Genetic algorithms are applied to solve inverse resonance-filter design problems. Applications including GMR laser mirrors, electro-optic modulators, and resonant Brewster filters are presented. Experimental results are shown to agree well with theoretical calculations.
The principles and chief properties of optical reflection and transmission filters based on guided-mode resonance (GMR) effects in multilayer structures comprising gratings and homogeneous thin films are presented. Detailed fiber characteristics (center wavelength, lineshape, and linewidth) are calculated using rigorous coupled-wave analysis for TE and TM polarized incident waves. These filters exhibit desirable characteristics such as high resonance efficiency with narrow or wide linewidths. Near- zero reflectance sidebands over extended wavelength ranges are obtainable using multilayer waveguide-grating structures. To illustrate the potential of this technology, calculated GMR reflection and transmission example characteristics are presented for filters made with common thin-film materials operating in the visible spectral region. Excellent reflection-filter features are found when antireflection conditions prevail away from the resonance wavelength. The transmission filter is optimized when the structure is highly reflective off resonance. It is found that long-range, low sidebands are obtainable for a single- layer GMR filter with a TM-polarized plane wave incident at the Brewster angle. GMR filter fabrication tolerances are briefly discussed. A calculated example illustrates the sensitivity of the filter center wavelength to variations in layer thickness. The effects of absorptive loss are treated. It is shown that, in general, GMR filters suffer loss- dependent wavelength shifts such that the reflection peak occurs at a different wavelength than the corresponding transmission notch. However, under antireflection conditions, the resonance location becomes insensitive to loss. Finally, reflective GMR thin-film structures that support multiple waveguide modes are studied. These devices exhibit characteristic angular and spectral signatures with unique appearance.
We present new, easy to use methods of measuring the fidelity of phase conjugated beams employing a classical system for image acquisition and processing, coupled to a personal computer. Two methods are based on intensity distribution changes measurements and the third is an interferometric evaluation of the wavefront reconstruction. These methods are applied in phase conjugation fidelity studies in a degenerate four wave mixing experiment using a He-Ne laser and a Bi<SUB>12</SUB>TiO<SUB>20</SUB> (BTO) photorefractive crystal. A comparison between the proposed methods, and also with other techniques currently used for phase conjugation fidelity studies, is made.