1) Field of the Invention
The present invention relates to an optical fiber applied to an optical amplifier suitable for optical communications and other related applications.
2) Description of the Related Art
In the field of recent optical communication systems, research and development in wavelength division multiplexing (WDM) transmission and building of optical communication networks using the WDM technology is now in progress. The WDM technology is inevitable for developing a large-capacity communication system.
An erbium-doped fiber (EDF), which is manufactured by doping element erbium (Er) into a core of a silica glass optical fiber, has an amplification band in a wavelength band of 1.55 micrometers that is the same as lowest loss band of the silica glass optical fiber. Furthermore, the EDF can take almost the same material and the same structure as an optical fiber for transmission, which makes it possible to realize a low connection loss when connecting with the optical fiber for transmission. For these reasons, the EDF is widely used as an optical amplification fiber for amplifying an optical signal in an optical transmission system.
An erbium-doped optical fiber amplifier (EDFA) employing the EDF has a high efficiency and a high gain, which can maintain a high-density pumping light and a signal light while overlapping the both in a long-haul transmission. The EDFA plays an important role as a key device in the WDM transmission system.
With the recent establishment of the WDM transmission technology, an introduction of the WDM transmission system became active in a long-haul large-capacity optical communication system, and a transmission capacity shows a continuing increase. Regarding such a WDM transmission, an experimental result of a transmission speed over terabit has been reported. When applying an optical amplification fiber, such as the EDF, in the long-haul high-speed large-capacity optical communication network, a high-quality optical amplification fiber is required.
One of the important parameters of a quality of amplification in the optical amplification fiber is the noise figure (NF). The noise characteristic of the optical amplification fiber is evaluated based on the NF.
One of the methods to reduce the NF is, as described in U.S. Pat. No. 6,804,046, to increase a ratio of an absorption coefficient αp in a pumping light wavelength band to an absorption coefficient αs in a signal light wavelength band. When the ratio αp/αs is increased, the ratio of the population inversion increases whereby the NF is decreased.
One of the refractive index profiles of a general optical amplification fiber is a step-index type, as shown in
However, in the conventional method disclosed in U.S. Pat. No. 6,804,046, it is necessary to narrow a rare-earth doped area in the center-doped type to increase the value of αp/αs. When the rare-earth doping area is narrowed, the absorption coefficient of the optical amplification fiber (absorption coefficient per unit length) becomes small. As a result, a length of the optical amplification fiber used must be longer to compensate for a decrease in the absorption coefficient.
Furthermore, in the center-doped type, it is difficult to match refractive indexes between the rare-earth doped area and a non-doped area in the core, resulting in a manufacturing problem.
It is an object of the present invention to solve at least the above problems in the conventional technology.
According to an aspect of the present invention, an optical amplification fiber for an optical fiber amplifier that amplifies a signal light using a pumping light, wherein an absorption coefficient of the optical amplification fiber satisfies conditional relations αs≧2.0 dB/m and αp/αs≧−0.0697×αs+1.30, where αp is the absorption coefficient in a wavelength band of the pumping light, and αs is the absorption coefficient in a wavelength band of the signal light.
According to another aspect of the present invention, an optical amplification fiber includes a core in which at least one rare-earth element is doped, the core having a first refractive index; a first cladding that surrounds the core, the first cladding having a second refractive index smaller than the first refractive index; a second cladding that surrounds the first cladding, the second cladding having a third refractive index bigger than the second refractive index and smaller than the first refractive index; and a third cladding that surrounds the second cladding, the third cladding having a fourth refractive index smaller than the third refractive index.
According to still another aspect of the present invention, an optical fiber amplifier includes the above optical amplification fiber.
According to still another aspect of the present invention, an optical communication system includes a transmitter that outputs a signal light; an optical fiber that propagates the signal light output from the transmitter; and the above optical fiber amplifier that amplifies the signal light using a pumping light.
The other objects, features, and advantages of the present invention are specifically set forth in or will become apparent from the following detailed description of the invention when read in conjunction with the accompanying drawings.
Exemplary embodiments of the present invention are explained in detail with reference to the accompanying drawings.
Any one of optical amplification fibers 11, 12, 13 according to the present invention can be applied to an optical fiber amplifier 10 shown in
The optical fiber amplifier 10 optically amplifies a signal light input, and outputs the amplified signal light. The optical fiber amplifier 10 includes an input terminal 16a that receives the signal light, an output terminal 16b that outputs the amplified signal light, an optical isolator 17a, 17b to true up a traveling direction of the signal light, a pumping light source 19a, 19b, a WDM coupler 18a, 18b to couple a pumping light from the pumping light source 19a, 19b to the signal light, and the optical amplification fiber 11 (or the optical fiber 12 or 13). The optical communication system 20 is for a long-haul transmission of the signal light, including a transmitter 21a that transmits the signal light, a receiver 21b that receives the signal light transmitted, an optical fiber transmission line 22 through which the signal light is transmitted, and the optical fiber amplifier 10 for amplification the signal light being transmitted.
First, a determination of a refractive index profile of the optical amplification fiber 11, 12, 13 is explained based on a numerical simulation, followed by an explanation for a performance evaluation for the optical amplification fiber 11, 12, 13 manufactured based on the refractive index profile determined by the numerical simulation.
To determine the refractive index profile of the optical amplification fiber 11, an absorption coefficient α(λ) is obtained from equality (1):
where ρ0 is Er concentration, ρ(r) is concentration distribution of rare-earth element, σa(λ) is absorption cross-section, (o is power mode radius, Ψ(r) is mode envelope, and a is an arbitrary value from among core diameters a1 to a3 shown in
Taking the EDF as an example, the power mode radius ω is obtained from the refractive index profile by using a numerical calculation. The power mode radius ω is then applied to the equality (2) to calculate the absorption coefficients α980 and α1530 at λ=980 nanometers and 1530 nanometers, respectively. Hereinafter, a pumping light wavelength and a signal light wavelength are set to 980 nanometers and 1530 nanometers, respectively. Although 980-nanometer-band and 1480-nanometer-band are generally used as the pumping light wavelength band for the EDF, it is known that a noise characteristic greatly depends on the pumping light wavelength band, and the pumping light in the 980-nanometer-band causes a lower noise than the 1480-nanometer-band. For this reason, the wavelength of 980 nanometers is used for the pumping light in the numerical simulation.
A first numerical simulation is carried out for the conventional step-index type (see
In an actual measurement, the Er-doped-area diameter d and the core diameter a in the first numerical simulation correspond to diameters at a position at which the refractive index becomes a half of the relative refractive index difference Δ1 shown in
The relative refractive index difference Δ1 is defined by:
Δ1={(nc1−nc)/nc1}·100 (3)
where nc1 is the maximum refractive index of the core, and nc is the refractive index of the cladding.
From the result of the first numerical simulation shown in
A second numerical simulation is performed to investigate a cut-off wavelength λc dependence of α980/α1530 in the W-seg type shown in
The cut-off wavelength λc dependence of α980/α1530 is investigated by fixing both a ratio of the core diameter a to the first cladding diameter b (Ra=a/b) and a ratio of the core diameter a to the second cladding diameter c (Rc=a/c) (Ra=0.8, Rc=0.48), and changing a ratio of a relative refractive index difference of the first cladding with the third cladding, Δ2, to the relative refractive index difference of the core with the third cladding Δ1(Δb=Δ2/Δ1) and a ratio of a relative refractive index difference of the second cladding with the third cladding, Δ3, to the relative refractive index difference of the core with the third cladding Δ1(Δc=Δ3/Δ1).
In this case, the relative refractive index difference of the core with the third cladding Δ1 is set to 1.0%, and the Er concentration is set to 1000 wt.ppm, which is uniformly doped in the core. The third cladding is substantially made of SiO2.
In a practical situation, the relative refractive index difference Δ1 and the relative refractive index difference Δ3 are corresponding to the maximum relative refractive index difference, and the relative refractive index difference Δ2 is corresponding to the minimum relative refractive index difference.
In the refractive index profile of the W-seg type shown in
The relative refractive index differences Δ2 and Δ3 are defined by the equalities (4) and (5). The equality (3) is used for the relative refractive index difference Δ1 with the refractive index of the third cladding as nc.
Δ2={(nc2−nc)/nc2}·100 (4)
Δ3={(nc3−nc)/nc3}·100 (5)
where nc2 is the minimum refractive index of the first cladding, nc3 is the maximum refractive index of the second cladding, and nc is the refractive index of the third cladding.
The result of the second numerical simulation shown in
However, if the cut-off wavelength λc is set on an extremely short wavelength side, the effective refractive index of the core becomes small, and as a result, a light confinement in the core becomes weak, and in the worst case, it cannot function as an optical fiber. For this reason, there is a limit in increasing α980/α1530 by setting the cut-off wavelength λc on the short wavelength side.
On the other hand, because α980/α1530 increases as Δc increases, a change of Δc, rather than a change of Δb, can cause a larger change of α980/α1530. In this manner, in the refractive index profile of the W-seg type, each of the relative refractive index differences Δ1, Δ2, and Δ3 can be adjusted as desired. Therefore, with an appropriate adjustment of each of the relative refractive index differences Δ1, Δ2, and Δ3 (for example, if Δc is adjusted to 0.36 or more), it is possible to obtain a larger α980/α1530 than that of the center-doped type even if the cut-off wavelength λc is the same.
A third numerical simulation is performed to calculate a change of α980/α1530 according to the ratio of the core diameter a to the first cladding diameter b (Ra=a/b) and the ratio of the core diameter a to the second cladding diameter c (Rc=a/c).
First of all, Rc and each of the relative refractive index differences Δ1, Δ2, and Δ3 are fixed (Δb=0.4, Δc=0.36, Δ1=1.0, Δ2=−0.4, Δ3=0.36, and Rc=0.36) to investigate the change of α980/α1530 according to Ra. The Er concentration is set to 1000 wt.ppm, which is uniformly doped in the core. The third cladding is substantially made of SiO2.
A further numerical simulation is performed to investigate a behavior of α980/α1530 according to the change of Ra when the cut-off wavelength λc is set to 900 nanometers. A result of the numerical simulation is shown in
For each of the samples, Rc and the relative refractive index differences Δ1, Δ2, and Δ3 are fixed, and the value of Ra is changed within a range between 0.4 and 0.9. The Er concentration is set to 1000 wt.ppm, which is uniformly doped in the core. The fixed values of Rc and the relative refractive index differences Δ1, Δ2, and Δ3 are shown in Table 1.
From the result shown in
A fourth numerical simulation is carried out to see a behavior of a change of Δ980/α1530 according to a change of Rc with Ra and the relative refractive index differences Δ1, Δ2, and Δ3 fixed (Δb=0.4, Δc=0.36, Δ1=1.0, Δ2=−0.4, Δ3=0.36, and Ra=0.9).
As shown in
A fifth numerical simulation is performed, based on the result of the fourth numerical simulation, to investigate the behavior of α980/α1530 according to the change of Rc when the cut-off wavelength λc is set to 900 nanometers, with Ra and the relative refractive index differences Δ1, Δ2, and Δ3 fixed (Δb=0.11, Δc=0.07, Δ1=1.8, Δ2=−0.2, Δ3=0.12, and Ra=0.7). The value of Rc is changed within a range between 0.14 and 0.5. The Er concentration is set to 1000 wt.ppm, which is uniformly doped in the core. A result of the fifth numerical simulation is shown in
From the result shown in
The absorption coefficient α1530 of the step-index type, when the cut-off wavelength λc is set to 900 nanometers, is about 5 dB/m to 6 dB/m, however, if Rc becomes 0.2 or less, the absorption coefficient α1530 shown in
To increase α980/α1530 in the conventional center-doped type, it is necessary to narrow the Er-doped area. However, if the Er-doped area is narrowed, the absorption coefficient α1530 decreases.
A sixth numerical simulation is carried out to investigate behaviors of α980/α1530 and the absorption coefficient α1530 in the W-seg type. At the same time, α980/α1530 and the absorption coefficients α1530 of the conventional center-doped type and step-index type are also simulated for a comparison.
For each of the step-index type, the center-doped type, and the W-seg-type optical amplification fibers, the cut-off wavelength λc is set to 900 nanometers, and the Er concentration is set to 1000 wt.ppm. The relative refractive index difference of the core Δ1 is set to a common value (Δ1=1.0). For the center-doped type, the Er− doped area is set to four levels: 0.2, 0.4, 0.6, and 0.8. For the W-seg type, Δb is set to a value within a range between 0.2 and 0.6, Δc is set to a value within a range between 0.12 and 0.48, Ra is set to a value within a range between 0.6 and 0.9, and Rc is set to a value within a range between 0.35 and 0.54. The cladding and the third cladding are substantially made of SiO2.
Based on a result of the sixth numerical simulation, behaviors of the absorption coefficient α1530 and α980/α1530 with each of the parameters (Δb, Δc, Ra, and Rc) for each of the refractive index profiles are plotted in a graph shown in
From the result shown in
A relation between α980/α1530 and the absorption coefficient α1530 in the center-doped type is obtained as α980/α1530=−0.0697×α1530+1.25 (a dashed line L2 in
α980/α1530>−0.0697×α1530+1.30 and α1530≧2.0 dB/m.
The results of the above first to sixth numerical simulations are verified by an actual measurement using the W-seg-type optical amplification fibers 11, 12, and 13. A result of the actual measurement is shown in Table 2. Results of the actual measurement for a conventional step-index-type optical amplification fiber 14 and a center-doped-type optical amplification fiber 15 are also described in the Table for a comparison.
An optical fiber having the W-seg-type refractive index profile can be practically manufactured by doping a dopant to increase the refractive index, such as Ge, in the core and the second cladding that have higher refractive indexes than the third cladding, and by doping a dopant to decrease the refractive index, such as F, in the first cladding that has a lower refractive index than the third cladding.
The refractive index can be adjusted by changing a doping amount of Ge or F. In general, the cladding and the third cladding do not include a dopant to change the refractive index, such as Ge and F, being made of substantially SiO2. Likewise, according to the present embodiment, the cladding and the third cladding are substantially made of SiO2. However, the present invention is not limited to the present embodiment.
The optical amplification fibers 14, 15, and the optical amplification fibers 11, 12, and 13 are manufactured with the cut-off wavelength λc set to about 900 nm. For the step-index-type optical amplification fiber 14 and the W-seg-type optical amplification fibers 11, 12, and 13, Er is doped in the entire core area. The Er− concentrations and Al− concentrations for the optical amplification fibers 14, 15, and the optical amplification fibers 11, 12, and 13 are substantially same and uniform.
A condition for measuring the NF, when actually measuring the NF of the optical amplification fibers 14, 15, and the optical amplification fibers 11, 12, and 13, is that a forward pumping light of 980 nanometers emitted from a laser diode (LD) is used, and a fiber length measured has a constant absorption coefficient-length product (absorption coefficient at 1530 nanometers×length).
From the result shown in Table 2, it is found that the W-seg-type optical amplification fibers 11, 12, and 13 have a larger α980/α1530 than the optical amplification fibers 14, 15 having the conventional refractive index profile. Furthermore, the W-seg-type optical amplification fibers 11, 12, and 13 can obtain an increased α980/α1530 without decreasing the absorption coefficient α1530, resulting in a great improvement of the NF.
According to the present invention, it is possible to increase a population inversion without accompanying a considerable decrease of the absorption coefficient in a wavelength band of a pumping light, which makes it possible to greatly decrease the NF.
Even if a rare-earth concentration and a cut-off wavelength applied to conventionally used step-index type and center-doped type are employed, a ratio of the absorption coefficient αp in the pumping light wavelength band (i.e., the wavelength band with a peak at around 980 nanometers) to the absorption coefficient αs in a signal light wavelength band (i.e., the wavelength band with a peak at around 1530 nanometers) can be increased, making it possible to decrease the NF. As a result, an optical amplification fiber with greatly improved noise characteristics can be realized.
Furthermore, because αp/αs can be increased without decreasing the absorption coefficient αs in the signal light wavelength, it is possible to make a length of the optical amplification fiber shorter, compared to a case in which the conventional step-index type or center-doped type is used. Therefore, a compact size of the optical fiber amplifier can be realized.
Furthermore, in an optical communication system employing an optical amplification fiber according to the present invention, an error rate on a reception side is decreased, realizing a long-haul transmission of a quality signal light.
Although the invention has been described with respect to a specific embodiment for a complete and clear disclosure, the appended claims are not to be thus limited but are to be construed as embodying all modifications and alternative constructions that may occur to one skilled in the art that fairly fall within the basic teaching herein set forth.
Number | Date | Country | Kind |
---|---|---|---|
2004-118544 | Apr 2004 | JP | national |