Optical waveguides – Directional optical modulation within an optical waveguide – Electro-optic
Reexamination Certificate
2000-04-28
2004-07-27
Lee, John D. (Department: 2874)
Optical waveguides
Directional optical modulation within an optical waveguide
Electro-optic
C385S037000, C385S123000, C398S147000, C398S148000
Reexamination Certificate
active
06768822
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to the compensation of chromatic dispersion, hereinafter referred to as dispersion, in optical transmission systems.
Linear (first order) dispersion, D, is the measure of the rate of change of group delay, &tgr;, with wavelength, &lgr;. Linear dispersion is typically measured in picoseconds per nanometre (ps
m). In the case of a transmission medium, for instance an optical fibre waveguide, whose waveguiding properties are uniform along its length, the linear dispersion exhibited by the medium is proportional to its length and so, for such a medium, it is convenient to define its linear dispersion per unit length, also known as its linear dispersion power. This is typically measured in picoseconds per nanometre per kilometer (ps
m/km).
The value of the linear dispersion of a transmission path is generally itself a function of wavelength, and so there is a quadratic (second order) dispersion term, Q, also known as dispersion slope, which is a measure of the rate of change of linear dispersion with wavelength. This is typically measured in picoseconds per nanometre squared (ps
m
2
).
BACKGROUND TO THE INVENTION
In a digital transmission system the presence of dispersion leads to pulse broadening, and hence to a curtailment of system reach before some form of pulse regeneration becomes necessary. The problem presented by dispersion increases rapidly with increasing bit rate. This is because, on the one hand, increasing the bit rate produces increased spectral broadening of the pulses, and hence increased dispersion mediated pulse broadening; while on the other hand, increasing the bit rate also produces a reduction in the time interval between consecutive bits. In a WDM (wavelength division multiplexed) digital transmission system, it is not practical to minimise the problems of dispersion by choosing to employ a transmission medium exhibiting near-zero first order dispersive power because low first order dispersive power is associated with aggravated non-linear (e.g. four-wave mixing) distortion. A known solution to this problem is to employ ‘managed dispersion’ in which near-zero aggregate linear dispersion over a particular transmission path is achieved by the use of alternating sections respectively exhibiting positive linear dispersion and negative linear dispersion, for instance by the use of NDS and DC optical fibre waveguide.
Having regard to the manufacturing tolerances in practice encountered in the fabrication of NDS and DC fibre, achieving adequately low aggregate linear dispersion becomes increasingly difficult as the bit rate is increased. Consider for instance a 40 Gbit/s WDM transmission system with a reach of 400 km, and with the shortest and longest wavelength channels separated by 200 nm. The actual amount of linear dispersion in any particular channel that can be tolerated at the receiver will of course be dependent upon a number of system parameters, but typically may lie in the region of 100 ps
m. A typical NDS fibre exhibits, at a wavelength of 1550 nm, a linear dispersive power of approximately 17 ps/(nmkm), and a quadratic dispersive power of approximately 0.058 ps/(nm
2
km). Currently DC fibre is fabricated to a tolerance of ±20% in respect of quadratic dispersive power. Therefore, for the 400 km span length, the uncertainty in linear dispersion compensation at the 1550 nm wavelength will amount to approximately 400 ps
m (400×17×0.06 ps
m).
Given the 200 nm wavelength range, the additional uncertainty at the wavelength extremities produced by the ±20% quadratic tolerance amounts approximately to a further 900 ps
m (400×0.058×200×0.2 ps
m). To this must be added any uncertainty arising from any imprecision in the knowledge of the length and dispersion of the transmission fibre.
The foregoing indicates that, even if the DC fibre were manufactured to tolerances tightened by an order of magnitude, those tolerances would still be large enough to cause difficulty in achieving an accurate enough compensation for the reliable provision of an operating point near the centre of the 100 ps
m window.
Additionally, in ultra long haul optical transmission systems (<3000 km) at 10 GBit/s or more, it is necessary to use high optical powers in the fibre in order to achieve sufficient signal to noise ratio at the receiver, A consequence of these high optical powers is that the nonlinear interactions caused by cross-phase modulation (XPM) and self-phase modulation are significantly enhanced. These two nonlinear mechanisms interact with chromatic dispersion to distort the pulse shape and cause timing jitter of the optical pulses. Whereas in a linear system it is possible to lump all the dispersion compensation in one place (e.g. at the receiver), in a nonlinear system, the distribution of the dispersion compensation can be critical. In particular it is necessary to control the net dispersion immediately following an optical amplifier, where the optical powers and hence nonlinearity are strongest. Simulations have shown the dispersion uncertainty at each amplifier site typically must be within 500 ps
m, which is the equivalent 30 RM of NDS fibre.
SUMMARY OF THE INVENTION
According to a first aspect of the present invention, an optical dispersion compensation device comprises a first optical compensation unit that applies non-linear dispersion compensation across a signal band, the first optical compensation unit being coupled to a second optical compensation unit that applies a degree of linear dispersion compensation across the signal band.
According to a second aspect of the present invention, a method of providing dispersion compensation comprises the steps of passing an optical signal through a first optical compensation unit which applies non-linear dispersion compensation across a signal band, and also passing the optical signal through a second optical compensation unit, coupled to the first optical compensation unit, which applies a degree of linear dispersion compensation across the signal band.
The approach taken in the present invention is to provide broadband dispersion compensation by applying dispersion slope compensation across the signal band to equalise residual dispersion slope and by applying a degree of linear compensation separately to affect the required linear dispersion compensation. Using these two degrees of freedom it is possible to set the desired dispersion slope and linear dispersion (whether positive or negative) to affect broadband dispersion compensation. Importantly, it is not necessary to demultiplex the optical signal.
In a preferred embodiment, the first optical compensation unit comprises an array of non-linearly chirped fibre gratings, each having a different dispersion slope, and an optical switch for selectively coupling one of the chirped fibre gratings into an optical path. Likewise, the second optical compensation unit comprises an array of linearly chirped fibre gratings, each offering a different degree of linear dispersion, and an optical switch for selectively coupling one of the linearly chirped fibre gratings into an optical path.
It is preferred that the chirped fibre gratings in the first optical compensation unit each have a relatively high dispersion slope with relatively low linear dispersion. Furthermore, the respective dispersion slopes in the array of chirped fibre gratings are centred about a pivotal point positioned substantially at the centre of the signal band.
In the preferred embodiment, a four port optical circulator is used to couple light between the first optical compensation unit and the second optical compensation unit, wherein one port is coupled to the first optical compensation unit and an adjacent port is coupled to the second optical compensation unit. The remaining ports of the optical circulator are coupled to a respective one of an optical input and an optical output of the optical dispersion compensation device.
Dispersion compensation components within the first and second optical compensation u
Fells Julian A.
Robinson Alan
Barnes & Thornburg LLP
Lee John D.
Nortel Networks Limited
LandOfFree
Chromatic dispersion compensation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Chromatic dispersion compensation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Chromatic dispersion compensation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3216204