This application claims priority to and the benefit of Korean Patent Application No. 2004-105709, filed Dec. 14, 2004, the disclosure of which is incorporated herein by reference in its entirety.
1. Field of the Invention
The present invention relates to a dispersion compensating fiber (DCF) for an S-band discrete Raman amplifier (RA).
2. Discussion of Related Art
Nowadays, as there have been intensive studies on the extension of the transmission band as part of a method for increasing transmission capacity, much attention has been paid to transmission techniques in the S-band, which is a low-loss band, as well as conventional C/L-band. Here, a loss characteristic of a silica fiber according to the wavelength is illustrated in
There are a rare-earth-element doped fiber amplifier such as a thulium doped fiber amplifier (TDFA), a Raman amplifier (RA) using Raman scattering in optical fibers, and a semiconductor amplifier, as amplifiers having gain in the S-band.
The TDFA employs thulium doped fibers (TDFs), which are made by doping thulium into nonsilica-based glass (i.e., fluoride glass), as amplification mediums. Thus, when the TDFA is connected to common silica fibers, great loss and problems are generated.
With the development in the manufacturing technology of semiconductor amplifiers, the semiconductor amplifiers are on the rise. However, a lot of technical limitations, for example, small gain and high polarization mode dispersion (PMD), still remain unsolved.
The RA has the advantage of having gain in various bands according to the pump wavelength, so it is appropriate for the S-band in which optical signals cannot be amplified using an EDFA. Above all, unlike a distributed RA that enables the Raman pumping of transmission lines, a discrete RA is inserted into a transmission system as an additional element to achieve gain. An S-band discrete RA includes pump laser diodes (LDs), optical fibers as amplification mediums, and other optical devices, such as a wavelength division multiplexer (WDM) that couples input optical signals and pump signals. So far, the gain of the S-band discrete RA has been controlled by adjusting the wavelength and output power of the pump LDs and the length of dispersion compensating fibers (DCFs). Accordingly, the S-band discrete RA cannot achieve high gain due to a limited output power of the pump LDs and the high loss of C-band DCFs and also has no adequate dispersion characteristics.
In this regard, a method of increasing gain by using nonlinear fibers having a high Raman gain coefficient along with conventional DCFs was introduced by David J. DiGiovanni [“Raman Amplified Dispersion Compensating Modules”, U.S. Pat. No. 6,504,973 B1, Jan. 7, 2003]. However, it is troublesome that the nonlinear fibers be designed and manufactured considering the dispersion of conventional DCFs, and it is difficult to expect a substantial increase in gain because the nonlinear fibers cause greater loss than the conventional DCFs.
The present invention is directed to a dispersion compensating fiber (DCF), which has dispersion characteristics appropriate for compensating dispersion in an S-band of single mode fiber (SMF) and a high Raman gain coefficient per effective area, thus improving the gain and dispersion characteristics of the S-band discrete RA.
In order to improve the gain and dispersion characteristics of the S-band discrete RA, the present invention provides the DCF, which has a higher Raman gain coefficient per effective area and a larger negative dispersion value than a conventional C-band DCF. Further, in order to efficiently compensate for the dispersion of a single mode fiber (SMF), the DCF is designed such that it has a dispersion slope opposite to that of the SMF.
One aspect of the present invention is to provide a DCF, which is an amplification medium of an S-band discrete RA. The DCF includes a core having a refractive index n1; an inner clad (or cladding) having a refractive index n2 and surrounding the core; and an outer clad having a refractive index n4 and surrounding the inner clad, wherein an ring having a refractive index n3 is disposed between the inner clad and the outer clad.
The core may have a trapezoid shape. Also, a difference in refractive index between the core and the outer clad may be 1.2 to 1.6%, a difference in refractive index between the inner clad and the outer clad may be 0.4 to 0.8%, and a difference in refractive index between the ring and the outer clad may be 0.2 to 0.6%. Further, the thickness of the ring may be 0.8 to 1.2 times as much as that of the core, and the thickness of the inner clad may be 1 to 3 times as much as that of the core. In the DCF, a Raman gain coefficient per effective area may be 6.5 km−1W−1 or more, an effective area may be 20 μm2 or less, and a dispersion value in the S-band may be −100 to −170 ps/nm/km. Preferably, a Raman gain coefficient per effective area is 6.5 km−1W−1 or more, an effective area is 9 to 11.9 μm2 or less, and a dispersion value in the S-band is −120 to −150 ps/nm/km.
The above and other features and advantages of the present invention will become more apparent to those of ordinary skill in the art by describing in detail exemplary embodiments thereof with reference to the attached drawings in which:
The present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments of the invention are shown. In the drawings, the thicknesses of layers may be exaggerated for clarity, and the same reference numerals are used to denote the same elements throughout the drawings.
In general, a Raman amplifier (RA) is based on the principle that gain is generated while an optical signal with high energy state is being converted into another signal with lower energy state by pump signal due to Raman scattering, which is a nonlinear phenomenon. The Raman gain spectrum of a pure silica fiber is illustrated in
The gain and output characteristics of the above-described S-band discrete RA are illustrated in
The DCFs used in the above-described S-band discrete RA are conventional C-band DCFs, and
Since the dispersion of the DCF is smaller in the S-band than in the C-band, when the dispersion of a single mode fiber (SMF) is compensated for as shown in
In this connection, the present invention proposes a new DCF, which has characteristics and structure appropriate for an amplification medium of an S-band discrete RA. The most suitable fiber for the S-band discrete RA should have the following characteristics:
1) A large negative dispersion value: A DCF should have a large negative dispersion value to compensate for the dispersion of an SMF in the S-band, and the length of the DCF required for the dispersion compensation should be small to minimize the loss of the DCF. Accordingly, in order to compensate for the dispersion of an SMF using a DCF with the smallest length, the DCF should have a large negative dispersion value. The dispersion of the C-band DCF in the wavelength of 1500 nm is −95 ps/nm/km.
2) An appropriate dispersion slope: In a transmission system, such as a wavelength division multiplexer (WDM) system, which simultaneously receives signals from 40 channels or more, the residual dispersion of each of the channels significantly affects the performance of the transmission system. The dispersion slope of the DCF is controlled such that the residual dispersion is minimized.
3) A high Raman gain coefficient: A Raman gain G is a function of Raman gain coefficient g, pump power P, effective area Aeff, fiber length L, and loss, as expressed in Equation 1:
where, α and α′ refer to loss in the wavelength of a pump signal and loss in the wavelength of an input signal, respectively, and κ refers to the polarization of an optical fiber having a cylindrical core (i.e., κ=2). Accordingly, the Raman gain coefficient g should be high to raise the Raman gain G.
4) A small effective area: Since the Raman gain of an optical fiber is inversely proportional to the effective area of the optical fiber, the effective area of the optical fiber should be scaled down to increase the Raman gain.
5) Low loss: In order to maximize the gain of an RA, loss should be small in the S-band. In a conventional C-band DCF, loss is 0.37 dB/km in the wavelength of 1500 nm as shown in
6) A suitable mode in a band: A fiber should be in a single mode in the S-band. Thus, cutoff wavelength should be less than 1300 nm.
A DCF that satisfies the characteristics 1) through 6) can be designed considering a relationship of the structure of the DCF with the dispersion and Raman gain coefficient thereof.
Meanwhile, a typical DCF has a staircase structure having a core with a high refractive index. However, because the staircase structure has a too high dispersion slope, the residual dispersion is great in a wide band. A structure for compensating for the dispersion slope of the staircase structure was proposed by Lars Gruner-Nielsen [“Dispersion-compensating module”, U.S. Pat. No. 6,654,531 B2, Nov. 25, 2003].
Also, the depressed clad structure can change the dispersion slope of a DCF, and a negative dispersion value can be larger by adding the ring to the depressed clad. The dispersion slope of a DCF should be opposite to that of an SMF, and the DCF should have a large negative dispersion value.
Accordingly, in the present invention, a DCF is designed as a structure including a core in which a rectangular portion and a triangular portion are properly combined, a depressed clad, and an outer clad with a ring.
Raman gain coefficient is proportional to the concentration of Ge doped into a core and the refractive index of the core, as expressed in Equation 2. Also, as the concentration of Ge in the core increases, the refractive index of the core increases. Thus, the Raman gain coefficient can be raised by elevating the refractive index of the core. However, a too high refractive index of the core leads to a small negative dispersion value of an optical fiber. Therefore, it is required to optimize the refractive index of the core in consideration of both the dispersion and the Raman gain coefficient.
wherein, σ0 refers to Raman cross sectional area and is proportional to the concentration of Ge as shown in Equation 3. C refers to a proportional constant, and XGeO2 refers to the concentration of Ge.
To reduce the effective area, the radius of the core should be minimized and the core should be designed in consideration of a dispersion characteristic. If the refractive index of the core is high, reflection occurs due to a large difference in refractive index between the core and the clad, and this leads to loss.
Accordingly, in the present invention, a ring is added to an outer clad for lowering loss caused by a large difference in refractive index between the core and the clad.
Also, the Raman gain coefficient of the DCF according to the present invention can be calculated using Equations 2 and 3. A variation in refractive index according to the concentration of Ge is obtained using the Sellmeier equation as shown in Equation 4:
wherein, SA, Sl, GA, and Gl refer to Sellmeier coefficients in SiO2 and Ge glasses, and X refers to the concentration of Ge.
Then, the effective area of the DCF according to the present invention is calculated using Equation 5. Thus, in the wavelength of 1500 nm, the effective area of the DCF is 11.9 μm2, which is considerably smaller than that (about 20 μm2) of the conventional C-band DCF.
Accordingly, if a value obtained by dividing the Raman gain coefficient g by the effective area Aeff is defined as Raman gain coefficient per effective area (g/Aeff), the g/Aeff of the DCF according to the present invention is 6.5 Km−1W−1, which is increased by about 47% more than that (4.4 km−1W−1 of the conventional C-band DCF.
Also, the ring is added to the outer clad in order to reduce loss caused by a difference in refractive index between the core and the clad. The cutoff wavelength is 700 nm, and the DCF is in a single mode in the S-band. The figure of merit (FOM) of the DCF is expressed in Equation 6. Here, the wavelength is 1500 nm, and the loss of the DCF according to the present invention is assumed as 0.8 dB/km considering the refractive index and radius of the core. Thus, the FOM of the DCF is 8.1 dB−1W−1, which is improved by about 62% as compared with the FOM (5 dB−1W−1) of the conventional C-band DCF.
FOM=g/(AeffEα) (6)
In conclusion, the DCF according to the present invention includes the trapezoid core having a small radius and a high refractive index, the depressed clad, and the outer clad with the ring. In this structure, the DCF, which is appropriate for the S-band discrete RA, has a large negative dispersion value in the S-band, a dispersion slope opposite to that of an SMF, and a high Raman gain coefficient per effective area so as to increase Raman gain.
As well known to those skilled in the art, in addition to the above-described method, there are a variety of methods for embodying optical fibers appropriate for an S-band discrete RA by making changes to the refractive indexes and radiuses of a core and a clad.
A DCF for an S-band discrete RA as proposed in the present invention has a large negative dispersion value in the S-band, a dispersion slope opposite to that of an SMF, a small residual dispersion, a high Raman gain coefficient per effective area, and a high FOM. Hence, the performance of the S-band discrete RA can be greatly enhanced.
Although exemplary embodiments of the present invention have been described with reference to the attached drawings, the present invention is not limited to these embodiments, and it should be appreciated to those skilled in the art that a variety of modifications and changes can be made without departing from the spirit and scope of the present invention.
Number | Date | Country | Kind |
---|---|---|---|
10-2004-0105709 | Dec 2004 | KR | national |
Number | Name | Date | Kind |
---|---|---|---|
3997241 | Nishida et al. | Dec 1976 | A |
5659649 | Nouchi et al. | Aug 1997 | A |
5675688 | Nouchi et al. | Oct 1997 | A |
6233387 | Berkey et al. | May 2001 | B1 |
RE37457 | Nouchi et al. | Nov 2001 | E |
6396987 | de Montmorillon et al. | May 2002 | B1 |
6400877 | Kato et al. | Jun 2002 | B1 |
6445864 | Jiang et al. | Sep 2002 | B1 |
6449416 | Lee et al. | Sep 2002 | B1 |
6504973 | DoGiovanni et al. | Jan 2003 | B1 |
6535676 | de Montmorillon et al. | Mar 2003 | B1 |
6633714 | de Montmorillon et al. | Oct 2003 | B1 |
6665481 | Hebgen et al. | Dec 2003 | B1 |
6751390 | Qi et al. | Jun 2004 | B1 |
6757468 | Bickham et al. | Jun 2004 | B1 |
6813430 | Aikawa et al. | Nov 2004 | B1 |
6819847 | Matsuo et al. | Nov 2004 | B1 |
6904217 | Kalish et al. | Jun 2005 | B1 |
6937805 | Aikawa et al. | Aug 2005 | B1 |
6952517 | Kalish et al. | Oct 2005 | B1 |
6999667 | Jang et al. | Feb 2006 | B1 |
20020003938 | Srikant | Jan 2002 | A1 |
20020182055 | Nagle | Dec 2002 | A1 |
20030081921 | Sillard et al. | May 2003 | A1 |
20030095769 | Aikawa et al. | May 2003 | A1 |
20030174987 | Bickham et al. | Sep 2003 | A1 |
20040146260 | Kalish et al. | Jul 2004 | A1 |
20040218881 | Rathje et al. | Nov 2004 | A1 |
20040234219 | Aikawa et al. | Nov 2004 | A1 |
20050063656 | Jablonowski et al. | Mar 2005 | A1 |
20050185905 | Burke et al. | Aug 2005 | A1 |
Number | Date | Country |
---|---|---|
10-2002 0029529 | Apr 2002 | KR |
20030003706 | Jan 2003 | KR |
Number | Date | Country | |
---|---|---|---|
20060126160 A1 | Jun 2006 | US |