Optical: systems and elements – Deflection using a moving element – Using a periodically moving element
Reexamination Certificate
1999-11-05
2002-08-27
Chan, Jason (Department: 2633)
Optical: systems and elements
Deflection using a moving element
Using a periodically moving element
C359S199200
Reexamination Certificate
active
06441939
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention concerns a system for regenerating a stream of solitons and finds an application in telecommunications networks.
Conveying data to be transmitted over long-haul telecommunications links in a stream of optical pulses is well known. A stream of this kind is timed, meaning that clock times are defined in regular sequence at a predetermined frequency constituting a “bit frequency”, the data to be transmitted being conveyed in binary form by the presence or absence of a pulse at each successive clock time. It is well known that the pulses can advantageously be launched into the fibers that waveguide them in the form of solitons. It is well known that a soliton is matched to the fiber which guides it and is then characterized by a specific amplitude time profile, a very short duration at half-height, and high spectral purity. Because chromatic dispersion and non-linearity effects (Kerr effect) specific to the fiber are compensated, a soliton has the advantage of propagating in the fiber without cumulative deformation, at least theoretically. However, the compensation requires that the average power of the pulse be maintained as it propagates, but power losses are inevitable and must therefore be compensated by means of a succession of optical amplifiers along the link. The amplifiers are typically optically pumped erbium-doped fibers. This need to compensate power losses of the pulses gives rise to two unwanted phenomena associated with spontaneous emission in the amplifiers: one is Gordon-Haus jitter, which consists in each soliton within the stream of solitons being displaced randomly on either side of the corresponding clock time. The other is accumulated noise. These two unwanted phenomena can degrade quality, in other words they can increase transmission error rate.
One prior art method for avoiding those drawbacks consists in synchronously modulating an optical pulse stream, more especially a stream of solitons. The modulation regenerates the stream of solitons. It restores the correct time profile and position of the solitons and eliminates or at least reduces noise. The theory of the method is described in an article by H. Kubota and Nakazawa, “Soliton transmission control in time and frequency domains”, IEEE J. Quantum Electronics v. 29, no. 7, pp. 2189-2197, July 1993. A 10 Gbit/s transmission experiment over one million kilometers is described in an article by Nakazawa et al. (1991), “Experimental demonstration of soliton data transmission over unlimited distance with soliton control in time and frequency domains”, Electronics Letters, v. 29, no. 9, pp. 729-730, Apr. 29, 1993.
A system for regenerating a stream of light pulses is described in an article by J. K. Lucek and K. Smith (1993) “All optical signal regenerator”, Opt. Lett. V. 18, no. 15, pp. 1226-1228, Aug. 1, 1993. Also worthy of mention is a communication by D. Sandel et al.: “Polarization-independent regenerator with nonlinear optoelectronic phase-locked loop”, Optical Fiber Conference Proceedings 1994, page FG2.
An article by M. Eiselt, W. Piper and H. G. Weber “SLALOM: Semiconductor Laser Amplifier in a Loop Mirror”, Journal of Lightwave Technology, vol. 13, no. 10, October 1995, pp. 2099-2112 describes a so-called SLALOM device. It indicates in particular that the device can be used for synchronous regeneration, as described in the section entitled: “SLALOM as Optical Retiming Device”. It does not specifically describe the essential components of a regenerator, i.e. a regeneration system, but rather those of an experimental system for assessing the possibilities of a regenerator. A regenerator as described in the above article is referred to hereinafter as a “prior art SLALOM regenerator”.
Some of the essential. features of a prior art regenerator of the above kind have analogies with features of a system of the present invention. These features are:
An optical waveguide between an input for receiving a data signal consisting of a stream of pulses to be regenerated and an output at which a stream of regenerated pulses carrying the same data is supplied. The pulses of the stream to be regenerated have an input wavelength. The stream has a clock rate defining successive clock times at a bit frequency. A segment of this waveguide constitutes an interferometer loop. The loop is closed by a loop coupler which couples the waveguide to itself at both ends of the loop to constitute a Sagnac interferometer.
A loop amplifier consisting of a semiconductor laser amplifier connected in series into the interferometer loop at a distance from the mid-point of the loop. This distance along the waveguide is referred to hereinafter as the “offset distance”.
Finally, a clock source which supplies an optical signal defining clock times corresponding to the stream of pulses to be regenerated. The signal is injected into the interferometer loop to bring about therein cross modulation between it and the pulses of the stream. Its wavelength is referred to as the clock wavelength.
In the prior art SLALOM regenerator, the clock source supplies the signal mentioned above in the form of clock pulses. The pulses are supplied to the input of the waveguide of the interferometer loop. The loop coupler derives components from the data signal pulses and the clock pulses. They circulate in the interferometer loop in two opposite directions. Those derived from the data signal cause temporary saturation of the loop amplifier. The offset distance mentioned above is chosen so that the pulses derived from the clock pulses reach the amplifier when it is in a saturated state or a non-saturated state depending on their time position relative to the previous pulses. If both components derived from a clock pulse reach the amplifier in two different saturation states, they interfere positively in the loop coupler and the clock pulse is therefore transmitted to the output of the waveguide of the interferometer loop. This situation arises if the data signal includes a pulse in a suitable time position relative to the clock pulse, i.e. if the data signal has opened a time window for the clock pulse. The system therefore behaves like a gate controlled by the data signal. In this way the data is transferred to the pulse stream consisting of the clock pulses transmitted to the output of the waveguide. The new data signal is therefore supplied at the clock wavelength.
It would also appear that in the above prior art regenerator the offset distance is defined as a function of the duration of the loop amplifier saturation state caused by each pulse of the signal initially carrying the data. This distance is therefore defined as a function of the lifetime of charge carriers in the amplifier.
The prior art SLALOM regenerator has the advantage of being of the “all-optical” kind, which avoids bandwidth limitations associated with the use of electronic signals. It would appear to have various other advantages compared to prior art modulators of other all-optical types. These advantages are in particular that its interferometer loop is much shorter than those of NOLM type systems and it does not require the use of polarization-maintaining fibers if the loop comprises an optical fiber and the modulator must be insensitive to polarization. These advantages are important because it is sometimes highly desirable to implement the modulator in integrated form and because an optical signal has random polarization when it travels great distances along a fiber optic link. However, the prior art SLALOM regenerator seems to have the particular disadvantage that its operation appears to be highly dependent on operating parameters that define the duration of the loop amplifier saturation state. Moreover, if it were applied to regenerating a stream of solitons, the synchronization signal would itself have to be in the form of a stream of solitons which could be described as “clock solitons”. Finally, it is necessary to allow for the fact that the wavelength of the new data signal is different from that of the signal received at the
Bigo Sebastien
Ducellier Thomas
Alcatel
Chan Jason
Leung Christina Y.
Sughrue & Mion, PLLC
LandOfFree
Device and method for regenerating a train of solitons does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Device and method for regenerating a train of solitons, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Device and method for regenerating a train of solitons will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2972682