JPS6148125B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an optical demultiplexer using a diffraction grating suitable for use in optical wavelength division multiplexing communications.

FIGS. 1 to 4 show examples of conventional optical demultiplexers using diffraction gratings, and their configurations and operating principles will be explained in advance, and then problems will be explained.

The optical demultiplexer shown in FIG. 1 was composed of a plane diffraction grating 1, a lens 2 as an optical coupling system, an input optical fiber 3, and a plurality of output optical fibers 4. The light from the input optical fiber 3 is made into parallel light by the lens 2, and this is collimated by the plane diffraction grating 1 at each wavelength λ ₁ , λ
₂ , λ ₃ ...... different angular dispersion (wavelength dispersion)
receive. The wavelength-dispersed light rays of each wavelength are converged again by the lens 2, and are then passed through the output optical fibers 4, 4' and 4', respectively, which have been arranged at different positions for each wavelength.
4'', 4, and operates as an optical demultiplexer.

The optical demultiplexer shown in FIG. 2 uses a spherical diffraction grating 5 instead of a plane diffraction grating, but since the spherical surface of the spherical diffraction grating 5 itself acts as an optical coupling system, the lens 2 in FIG. Omitted.
The operating principle is similar to that of FIG. However, α
is the angle of incidence with respect to the optical axis of the spherical surface.

The optical demultiplexer shown in FIG. 3 uses a cylindrical diffraction grating 6 instead of a plane diffraction grating, and a flat optical waveguide 8 having a high refractive index layer 7 with a uniform refractive index distribution as an optical coupling system.
It uses That is, for a light beam that has undergone wavelength dispersion in the cylindrical surface diffraction grating 6, convergence in the wavelength dispersion direction is performed by the cylindrical surface of the diffraction grating 6 itself, and convergence in the direction perpendicular to the wavelength dispersion is performed by the flat optical waveguide 8. The operating principle is similar to that of FIG.

The optical demultiplexer in FIG. 4 uses a plane diffraction grating 1 instead of the cylindrical diffraction grating 6 in FIG.
As an optical coupling system, the refractive index is 2 in the wavelength dispersion direction.
A flat optical waveguide 8' is used which has a high refractive index layer 7' having a power distribution and a uniform refractive index distribution in the direction perpendicular to the wavelength dispersion. In this case, the wavelength-dispersed light beam is converged in each direction only by the flat plate optical waveguide 8', but the operating principle is the same as that in FIG. 1.

By the way, in the conventional optical demultiplexer described above, (1) the direction of the reflective surface of the grooves of the planar diffraction grating 1 and the groove interval are uniform over the entire surface of the diffraction grating, and (2) the spherical diffraction grating 5 and the cylindrical surface In the diffraction grating 6, although the groove spacing was continuously changing, the reflecting surfaces of the grooves were oriented in the same direction and did not change continuously.

Therefore, conventional optical demultiplexers have had problems such as large losses. That is, in the ones in FIGS. 1 and 4, the optical coupling system 2,
8' causes the image of the output light to become blurred compared to the image of the input light, resulting in deterioration of the coupling between the output light and the output optical fiber 4, or the wavelength that should be coupled to one of the output optical fibers 4. A phenomenon occurs in which a portion of the output light of one optical fiber is coupled to another output optical fiber, resulting in problems such as increased loss in the optical demultiplexer and increased mutual crosstalk between channels. In order to improve this, conventionally, a set of lenses or a flat optical waveguide 8' has been used.
Measures were taken to reduce aberrations in the optical coupling system, such as bringing the refractive index distribution of the high refractive index layer 7' closer to the ideal value.
However, these measures were not sufficient due to problems such as limitations in the degree of improvement and high costs.

On the other hand, in the systems shown in Figures 2 and 3, the aberrations of the optical coupling system are corrected by continuously changing the groove spacing of the spherical diffraction grating 5 and the cylindrical diffraction grating 6, but the reflective surfaces of the grooves are continuous. Since the groove spacing does not change, deterioration in diffraction efficiency is unavoidable, resulting in an increase in loss in the optical demultiplexer.Also, there is a problem in that mutual crosstalk between channels cannot be avoided simply by changing the groove spacing.

Note that the above explanation is about an optical demultiplexer, but if the input/output relationship is reversed, it becomes an optical multiplexer, and even in the case of an optical multiplexer, there is a problem of increased loss for the same reason.

SUMMARY OF THE INVENTION In view of the above-mentioned problems of the prior art, it is an object of the present invention to provide an optical demultiplexer having good characteristics with low optical insertion loss and low crosstalk between channels. This objective can be achieved by continuously changing the reflection surface of the grooves of the diffraction grating and the interval between the grooves over the entire surface of the diffraction grating. The grooves of such a diffraction grating can be easily manufactured, and the optical demultiplexer can be manufactured at low cost. The present invention will be described below with reference to the drawings.

First, as a representative example of the present invention, an optical demultiplexer using the cylindrical surface diffraction grating 6 shown in FIG. 3 will be described. FIG. 5 shows a configuration diagram in this case. However, illustration of the flat optical waveguide 8 is omitted. In FIG. 5, the groove angle θ with respect to the tangent of each part of the cylindrical surface of the cylindrical diffraction grating 6 is determined by the reflection surface of each groove being connected to the input optical fiber 3.
The grooves are set so that they always form a right angle to the straight line connecting the center point on the light output end face of the groove and each groove. Further, the groove spacing dx is set by continuously changing it over the entire lattice surface according to the following equation (1).

However, dx is the groove spacing, do is the groove spacing at the center of the cylindrical diffraction grating 6, θ is the groove angle, θ ^f _x is the exit angle of the light beam emitted from the input optical fiber 3, and θ ^g _x is the groove angle θ and the exit This is an angle given by the following equation (2) in relation to the angle θ ^f _x .

θ ^g _x = θ+sin ^-1 {−sinθ・cos ² θ ^f _x +sinθ ^f _x √1− ²・^{2 f} _x }...
...(2) Set the groove angle θ as described above and calculate the groove distance dx using the formula
As shown in (1), when the input ray is varied as a function of the output angle θ ^f _x , since the diffraction grating is on a cylindrical surface, it is diffracted from each groove at the same angle to the reflecting surface. A light beam of a certain wavelength converges at one point corresponding to the focal point of the cylindrical surface, and is input to the output optical fiber 4' whose light input end face is placed at this converging point. The position corresponding to this focal point differs depending on the wavelength. In such a groove configuration, the effect of continuously changing the groove spacing is compared with the case where the groove spacing is constant.
As shown in the figure. However, in FIG. 6, the horizontal axis indicates the exit angle θ ^f _x (degrees) of the light beam emitted from the input optical fiber 3, and the vertical axis indicates the end face of the output optical fiber 4' after a certain light beam 9' in FIG. 5 is diffracted. The distance (displacement) △x between the position where the other light rays, for example 9'', 9, are diffracted and incident on the end face of the output optical fiber 4'
It is.

In FIG. 6, the solid line shows the case where the groove spacing is continuously changed, the broken line shows the case where the groove spacing is constant,
According to the configuration of the present invention, Δx is zero regardless of the angle θ ^f _x of the output beam of the input optical fiber 3, and all the beams converge at approximately one point on the end face of the output optical fiber with corresponding wavelengths. In addition, in the case of a conventional groove configuration in which the groove spacing is constant, the position shifts by about 33 μm at the maximum depending on the output angle θ ^f _x from the input optical fiber 3.
In other words, in the past, after the light rays emitted from a point light source received the diffraction effect and convergence effect on the cylindrical diffraction grating,
The diffraction image does not converge to a single point, resulting in a greatly blurred diffraction image (aberration). This means that in an optical demultiplexer, a portion of the light beam of a wavelength that should entirely enter a certain output optical fiber, e.g. 4', leaks into other output optical fibers, e.g. 4'', 4; The mutual crosstalk characteristics between channels of the optical demultiplexer will deteriorate.

When a groove configuration in which the groove spacing is continuously changed as in the present invention is used, the above-mentioned blurring of the conventional diffraction image is completely eliminated, and therefore the mutual crosstalk characteristics between the channels of the optical demultiplexer are not deteriorated. .

On the other hand, in FIG. 5, the angle between the tangent to the cylindrical surface of the diffraction grating 6 and the reflective surface of the groove, that is, the groove angle θ, is calculated using the following equation (3).
If the settings are as follows, the diffraction grating 6 will be connected to the optical fiber 3.
Since the incident light ray satisfies a certain condition for diffraction generally called the blaze condition, the m-th order diffracted light near the wavelength λ _B of the reflected diffracted light becomes the strongest, and as a result, it is input to the output optical fiber 4'. Almost all of the light rays are of diffraction order m.

θ=sin ^-1 m・λ _B・δ _x /2R・δ〓……(3) However, λ _B is the wavelength that we are currently focusing on, which is called the blaze wavelength, and δ _x is the spacing between the output optical fibers, R is the curvature of the cylindrical diffraction grating, δ〓 is the center wavelength spacing of adjacent channels, and m is an integer.

This configuration in which the direction of the reflective surface continuously changes depending on the groove angle θ and the conventional configuration, that is, a configuration in which the entire surface of the cylindrical diffraction grating is divided and the direction of the reflective surface of the groove is constant for each divided portion, are different from each other. The difference in efficiency is the seventh
As shown in the figure.

In FIG. 7, the solid line indicates the diffraction efficiency when the groove angle based on equation (3) is adopted, and as shown on the left vertical axis, a high diffraction efficiency of 99.99% or more is obtained over the entire surface of the diffraction grating. On the other hand, the broken line shows the diffraction efficiency of the conventional three-division system, and as shown on the right vertical axis, the diffraction efficiency is 100% at the center of each division, but the diffraction efficiency deteriorates to less than 50% at the joints of each division. and
When averaged, the diffraction efficiency is only about 70% at most. In other words, considering the characteristics of the optical demultiplexer, the insertion loss of the optical demultiplexer is improved by about 30% and reduced by using the groove angle θ in equation (3).

The optical demultiplexer using the cylindrical diffraction grating with the groove configuration shown in FIG. 5 as described above has good characteristics with lower loss and less mutual crosstalk between channels compared to conventional optical demultiplexers. It will be done. Conversely, when the device shown in FIG. 5 is used as an optical multiplexer, light rays of any wavelength converge at one point on the end face of one optical fiber 3, and the diffraction efficiency is high, resulting in extremely low loss.

By the way, the above explanation is about an example in which the present invention is applied to the optical demultiplexer using the cylindrical diffraction grating shown in FIG. This can be applied to optical demultiplexers, and even more so to optical demultiplexers using the plane diffraction grating 1 shown in FIGS. 1 and 4. That is, when the present invention is applied to the optical demultiplexer shown in FIG.
Since the structure is similar to that shown in the figure, similar characteristics can be obtained by providing the groove structure explained in FIG. 5 to each cross section. In addition, in the case of the optical demultiplexer shown in FIGS. 1 and 4, the groove configuration, that is, the groove angle It is easily seen that the same effect as the cylindrical diffraction grating described above can be obtained by setting and continuously changing the groove spacing.

Furthermore, it is also possible to use an optical demultiplexer with a plurality of input optical fibers 3, or an optical demultiplexer with a configuration in which a receiver is provided in place of the output optical fiber 4 and the demultiplexed light is immediately converted into an electrical signal and taken out. .

As explained in detail above, in the optical demultiplexer using the diffraction grating according to the present invention, by setting the groove angle of the diffraction grating used therein and continuously changing the groove spacing over the entire grating surface, the optical demultiplexer can be The aberrations of the optical coupling system can be reduced, and mutual crosstalk between channels can be reduced.
Moreover, diffraction efficiency is improved and insertion loss is reduced.

[Brief explanation of the drawing]

1 to 4 are configuration diagrams showing examples of optical demultiplexers using diffraction gratings, and FIG. 5 is a configuration of an embodiment in which the present invention is applied to the optical demultiplexer shown in FIG. 3. Figure, 6th
The figure is a graph showing the effect of reducing aberrations, and FIG. 7 is a graph showing the effect of improving diffraction efficiency. In the drawing, 1 is a plane diffraction grating, 2 is a lens, 3 is an input optical fiber, 4 is an output optical fiber, 5 is a spherical diffraction grating, 6 is a cylindrical diffraction grating, 7 and 7' are high refractive index layers, 8 and 8' is a flat plate optical waveguide.

Claims (1)

[Scope of Claims] 1. At least one input optical fiber, at least one light-receiving element such as an optical fiber or a light receiver, and at least one optical coupling system that optically couples the input optical fiber and the light-receiving element. , an optical demultiplexer consisting of a cylindrical grating that diffracts input light of multiple wavelengths from an input optical fiber at different angles for each wavelength, where the cylindrical grating covers the entire surface of the grating, and θ= sin ^-1 m・β _B・δ _x /2R・δ〓 However, m is the diffraction order, λ _B is the brace wavelength, δ _x is the spacing between the light receiving elements, R is the curvature of the cylindrical diffraction grating, and δ〓 is This is the center wavelength spacing of adjacent channels. The groove angle θ given to the tangent to the cylindrical surface of the grooved part by the formula, However, do is the groove spacing at the center of the cylindrical diffraction grating, θ ^f _x is the exit angle of the light beam emitted from the input optical fiber, and θ ^g _x is the angle given by θ and θ ^f _x . An optical demultiplexer comprising a continuously changing groove spacing d _x given by the following formula. 2. At least one input optical fiber, at least one light receiving element such as an optical fiber or a light receiver, and a spherical diffraction grating that diffracts input light of a plurality of wavelengths from the input optical fiber at a different angle for each wavelength. In the optical demultiplexer, the spherical diffraction grating has the following formula in all cross sections including the optical axis: θ=sin ^-1 m・λ _B δ _x /2R・δ〓 where m is the diffraction order and λ _B is the blaze The wavelength, _δx , is the spacing between the light receiving elements, R is the curvature of the spherical diffraction grating, and δ〓 is the center wavelength spacing of adjacent channels. The groove angle θ given to the tangent to the spherical surface of the grooved part by the formula, However, do is the groove spacing at the center of the spherical diffraction grating, θ ^f _x is the exit angle of the light beam emitted from the input optical fiber, and θ ^g _x is the angle given by θ and θ ^f _x . An optical demultiplexer comprising a continuously changing groove spacing dx given by the following formula.