A 4-dB reduction of the optical signal-to-noise ratio admissible by an all-optical regenerator is demonstrated. Applied to long-haul transmission, an error-free 10-Gbit/s NRZ transmission over 170,000 is performed with 170-km repeater spacing.

We present for the first time a physical validation of an all-optical packet-switched network. More than 40 network sections, including 100 km of fibre and one optical packet switching node, have been cascaded with negligible sensitivity penalty at 10 Gbit/s.

Summary form only given. The large scale deployment of all-optical wavelength-division multiplexing (WDM) networks will require signal regeneration due to signal impairments occurring in transmission spans and optical switching nodes/cross-connects. A 2R regenerative interface for non-return to zero (NRZ) signals based on semiconductor optical amplifiers (SOAs) has been proposed by Chiaroni et al. (1997), with clear demonstration of regenerative properties in the amplitude domain. In this paper, a 3R regenerative interface based on the same core structure is described. The core of these interfaces is composed of the association of a wavelength converter (WC) under cross-gain modulation (XGM-WC) and a wavelength converter under cross-phase modulation (XPM-WC), both operating in polarity-inverting mode. Basically, the role of the first WC is to convert power fluctuations into extinction ratio variations and to provide the second WC with a constant input power. The second converter is in turn mainly used to regenerate the extinction ratio thanks to its nonlinear transfer function. This interface in addition limits the accumulation of amplified spontaneous emission (ASE) noise.

Performances in terms of minimum admissible signal-to-ASE ratio and power dynamic range and system margins of a semiconductor optical amplifier-based regenerator have been investigated at 10 Gbit/s with NRZ signals in long haul transmission experiments.

The large growth of telecommunication traffic demand generated by multiple new applications and expected to last at least for the next decade will force telecom operators to consider offering more flexible transport services. All-optical packet switching is a powerful technique to provide this flexibility and to support in a cost-efficient way a wide range of bandwidth consuming applications. After a very brief introduction about the packet-switched network architecture studied in the framework of the ACTS KEOPS project, we describe the structure of the packet-switching node we have defined. We then move into physical and logical analysis of the network including more than 40 network sections based on 160 Gb/s throughput optical packet switching nodes could operate error free. In addition, logical simulations have proved that such networks could provide a quality of service (packet loss rate and packet transfer delay per node) compatible with a large variety of service classes. Both results validate the feasibility of the network concept and pace the way toward a flexible network based on all-optical switching techniques.

One of the objectives of the the European ACTS 043 KEOPS project, was to assess the feasibility of a high capacity all-optical packet switching network to face the dramatic increase of traffic needs. The initial objective was to cascade a maximum of 16 network sections (involving transmission links up to 100 lan and one optical packet switching node) at 10 Gbit/s to validate the concept. In this paper we present both the experimental validation and a logical analysis. The physical performance has been assessed through a loop cascade of 40 network sections including 160 Gbit/s throughput switching nodes and 100 lan or transmission. Recent experimental results have shown that such a network could be extended to a world scale. The limits of operation have been checked by regarding interferometric noise influence in the cascade and evolution of power discrepancies through the network. For the first time, these results really indicate that it is possible to provide high capacity, full flexibility and total expandability at the network level without any opto-electronic conversion. Finally, we will give simulation results exhibiting the packet loss rate, the packet delay and the occupation rate in the buffer. In particular, we demonstrate that the packet loss rate was preserved during the cascade. Results are compatible with ATM constraints as well as with other data transmission formats. This set of results demonstrates the feasibility of an all-optical packet switched network while providing both high quality of signal and high traffic performance.

The future telecommunication network will have to face the dramatic increase of subscribers as well as the increase of the user bandwidth through new services. All-optical packet switching techniques can become a strategic objective to offer on an unique technology a service-transparent network. In this paper, we will describe in detail the structure of an optical packet switching node developed in the framework of the ACTS 043 KEOPS project. An analysis of the key functions will be reported to fulfil. system requirements including cascadability. In particular the input synchronisation, the Broadcast-and-select switching matrix and the output regenerative interface will be described and physical performance will be assessed through theoretical analysis : quality of the signal, packet jitter and packet power fluctuation. The electronic circuitry for the control of the components of each sub-block will be described. Finally, experimental validations of a 160 Gbit/s throughput node will be reported. In order to complete the analysis, the logical performance in a Bernoulli-type traffic will be regarded. In particular an optimised buffer including a recirculation loop will be studied. Logical performance exhibiting a packet loss rate lower than 10(-9) for a 0.8 load and mean packet delay as low as 3 packet slots will be illustrated, thereby demonstrating full compatibility with ATM constraints. Finally, new perspectives in terms of throughput potential through cascading will be drawn.

A new NRZ optical semiconductor optical amplifier-based 3R signal regenerator structure with high integration potential is assessed through loop experiments. More than 50 regenerators have been cascaded nearly penalty-free at 10 Gbit/s, and regeneration properties are highlighted

Wavelength Division Multiplexing (WDM) is a technique to upgrade installed lightwave systems and allows for optical networking functionalities. The requirements on stability of light sources, filters, and other optical components is such that a channel spacing of 200 GHz is emerging as a viable solution. In this context, increasing the throughput leads to an increase of either the single-channel bit-rate or the total occupied bandwidth. In the first case, the achievable distance will quickly become limited by the fiber polarization mode dispersion (PMD), decreasing as the square of the bit-rate. The second case leads to an increase of the number of channels with the use of flat-gain optical amplifiers.

16 10 Gbit/s channels spanning 24 nm were transmitted over 531 km of dispersive fibre with only seven fluoride-based EDFAs, amplifier spacing varied between 60 and 93 km, and optimised dispersion management using dispersion-compensating fibre was utilised.

We develop a simple method leading to the determination of important parameters for the praseodymium-doped fluoride fibre amplifier (PDFFA), such as the non-radiative energy transfer upconversion coefficient or pump-excited state absorption cross section . We apply this method to five ZBLAN bulks of different concentration (from 10 000 down to 500 wt ppm) and to a ZBLAN: 500 wt ppm single-mode fibre. The resulting non-radiative energy transfer upconversion coefficients are at 500 wt ppm and at 1000 wt ppm. When the pump wavelength is set at 1017 nm, is very weak: less than . We measured that the population varies according to the pump power raised to the power 2.7. We conclude that the upconversion mechanism leading to the excitation of states in the PDFFA is different from all that which has been previously suggested and we propose a three-photon process. The application of our method requires the measurement of radiative emission branching ratios from levels and that we compare to the values calculated with the Judd - Ofelt analysis. We show that this analysis has to be used very cautiously with ions. Finally, we discuss the distribution of ions in ZBLAN glass. The fitting of fluorescence decay curves from level with the Inokuti-Hirayama model shows that the distribution of ions in ZBLAN is nearly homogeneous and no clustering effects are detected. Nevertheless, the non-exponential fluorescence decay curve of level even at very low concentrations is explained by the fact that the ions do not all have exactly the same environment in the glass.

The flat gain over a broad spectral width offered by fluoride-based Erbium-Doped Fiber Amplifiers (EDFA) appears very attractive for wavelength-multiplex systems operating near 1550 nm. The relevant parameters (input pump and signal powers, fiber length) that determine the gain flatness characteristics of fluoride-based EDFA have been experimentally investigated. It is found that flat gain operation requires an optimal, though not critical, population inversion within the doped fiber, which is somewhat lower than the maximum population inversion achievable with 1.48 mum-pumping. Implications on the design and operating conditions of fluoride-based EDFA are discussed.

With the ever growing interest in Wavelength Division Multiplexing transmission systems, the need for an optical amplifier with a large bandwidth is more and more necessary. Unlike the standard Erbium-doped silica fiber amptifiers, the fluoride-based fiber amplifiers display more uniform gain spectra. The benefit expected from fluoride amplifiers with multiwavelength signals has been experimentally assessed through comparisons in single- and cascaded-amplifiers configurations. 4 dB gain flatness was achieved with four wavelength-multiplexed signals extending over 23 nm after three fluoride amplifiers while 15 dB gain unbalance was observed after three silica amplifiers. This result shows that fluoride amplifiers may be of the utmost relevance for multiwavelength transport systems.

A new technique for flat gain control over a large range of input power with 1.55 mu m fluoride-based erbium-doped fibre amplifiers is reported. It relies on the monitoring of backward-propagating amplified spontaneous emission. This technique is promising for future multiwavelength systems, where signal power or channel number variations may occur.<>

Using an 8 channel multiplex ranging over 28 nm and a circulating loop, the gain excursion and the signal-to-noise ratio discrepancies between channels were measured against the number of in-line fluoride-based fibre amplifiers. 27 nm 6 dB gain bandwidth was observed after six fluoride-based amplifiers.<>

While optical fiber medium offers a THz-wide bandwidth, silica-based Erbium-Doped Fiber Amplifiers (EDFA) represent a bottleneck in broadband all-optical lightwave systems. This is due to their spectral gain ripples and non-uniformities, resulting in gain and Signal-to-Noise Ratio (SNR) discrepancies between channels. Different techniques have been proposed to overcome this impediment [1-4]. However, they have never been demonstrated over a spectral range larger than 14 nm while EDFAs offer more than 20-nm operating bandwidth. Based upon a different glass material, Erbium-Doped Fluoride Fiber Amplifiers (EDFFA) exhibit gain spectra flat and more uniform than silica-based EDFAs [5,6]. In the experiment reported here, this gain flatness advantage is used over 25 nm with 16 wavelength-multiplexed channels for cascade operation in a 440 km-long system. A transmission experiment demonstrates a potential of 40-Gbit/s total capacity through 440 km standard fiber, using fluoride-based EDFAs and a 16-channel multiplex.

The feasibility of 40 Gbit/s aggregate bit rate transmission through 440 km standard fibre is experimentally demonstrated using a 25 nm-wide 16 wavelength multiplex and fluoride-based fibre amplifiers. This result corresponds to a potential bit rate-distance product of 17.6 Tbit/s.km through standard fibre.

Within lightwave analog amplitude-modulated (AM) CATV systems using directly modulated lasers, erbium-doped fiber amplifiers (EDFA's) act upon the signal distortion because of the interaction between the laser chirp and the EDFA wavelength-dependent gain. This interaction is theoretically investigated in order to predict the EDFA-induced distortion. The relevant gain tilt characteristic for analog applications and the way to measure it are described. Expected and measured distortions at the EDFA output are in excellent agreement. Fiber amplifiers are found to decrease the signal distortion level when the gain tilt is negative, i.e., for wavelength above the gain maximum.< >