JP5910970B2 - Wavelength selective switch - Google Patents

Description

  The present invention relates to a wavelength selective switch introduced into a wavelength division multiplexing optical communication line.

  In a communication node installed in a wavelength division multiplexing (WDM) communication path, an optical signal is converted into an electrical signal and the path is switched by an electrical switch, and the path is switched without changing the optical signal. Introduction of a reconfigurable optical add / drop multiplexer (ROADM) system capable of simultaneously realizing low power consumption and miniaturization of a communication node is in progress. The ROADM is equipped with a wavelength selective switch (WSS) as a core device, and various methods have been proposed so far.

FIG. 1 is a conceptual diagram of a wavelength selective switch having a general configuration, and its operation will be described. A general WSS includes an optical input / output unit having at least one input port and N output ports, at least one lens, a wavelength demultiplexing unit and a wavelength multiplexing unit (wavelength multiplexing / demultiplexing unit), and an optical signal. And a light deflection unit for switching to a desired output port. FIG. 1 shows an example in which the optical input / output unit 100 is configured by an input / output fiber array 101, the wavelength multiplexing / demultiplexing unit is configured by a diffraction grating 105, and the optical deflection unit is configured by an optical deflection element 107 in the configuration of the wavelength selective switch. .

  In the following, in the wavelength selective switch, the direction of wavelength demultiplexing by the wavelength multiplexing / demultiplexing unit is the x axis, the direction in which the fibers of the input / output fiber array 101 that is orthogonal to the x axis are arranged is the y axis, and the optical signal is The traveling direction when output from the fiber is defined as the z-axis.

  In the following, for the sake of explanation, the number of the input / output fiber arrays 101 constituting the optical input / output unit 100 is set to five, the central fiber is set as the input port, and the fiber arranged at the bottom is set as the output port. The selection of the number and input / output fibers is not limited to this description. Further, the chief ray through which the optical signal emitted from the input fiber passes to the optical deflecting element 107 is shown by a thick solid line, and the chief ray until the optical signal reflected by the optical deflecting element 107 is coupled to the output fiber is thick. It is represented by a broken line.

  An optical signal emitted into the space from the input / output fiber array 101 propagates while spreading at a constant numerical aperture (NA) corresponding to the beam diameter confined by the input / output fiber array 101. The NA of this optical signal is adjusted by the microlens array 102 whose focal length and arrangement position are adjusted so that the optical signal emitted from each optical fiber of the input / output fiber array 101 propagates through space as collimated light. Further, it is converted into an angle corresponding to the incident position of the condenser lens 103 in the y-axis direction. This optical signal is converted into an angle again through the lens 104, and then wavelength-demultiplexed in the x-axis direction by the diffraction grating 105, and further condensed on the optical deflection element 107 through the lens 106. In the optical deflection element 107, the optical signal whose angle is adjusted so as to be optimally coupled to the output fiber passes through each lens and the diffraction grating in the reverse direction and reaches the condenser lens 103. The condensing lens 103 has a function of converting to an angle proportional to the height of the beam incident on the condensing lens 103 in the y-axis direction, and an optical signal incident on the condensing lens 103 is transmitted to the output fiber. The position and angle are adjusted so as to be parallel to the z-axis, and switching can be performed by coupling to a desired output fiber in the input / output fiber array 101. Since this operation is wavelength-demultiplexed by the diffraction grating 105, it can be distributed to a desired output port for each wavelength by setting the optical deflecting element 107, and operates as a wavelength-selective switching device.

  As the optical deflecting element 107 constituting the optical deflecting unit, a micromirror array (for example, refer to Patent Document 1) based on MEMS (Micro-electro mechanical system) technology, a liquid crystal cell array, a DMD (Digital mirror device), an LCOS (Liquid crystal on silicon) (for example, refer to Patent Document 2).

  When realizing WSS, the basic operation principle depends on what elements are used to configure the input / output unit, wavelength multiplexing / demultiplexing unit, and optical deflection unit, which are the above-described important components. The features will be very different.

  For example, as the simplest configuration, WSS is known in which an input / output unit is a fiber array, a wavelength multiplexing / demultiplexing unit is a bulk diffraction grating, and an optical deflection unit is LCOS. In this example, optical elements are arranged in substantially the same manner as in FIG. As a feature of the configuration example of the WSS, since the optical fiber array 101 is employed for the input / output unit, the input / output ports can be easily arrayed, and the number N of output ports can be set to a large value of about several tens. In addition, since many bulk optical elements are used, a relatively low loss WSS can be realized.

  As another WSS configuration example, both the input / output unit and the wavelength multiplexing / demultiplexing unit are integrated in an arrayed-waveguide grating (AWG) using a planar lightwave circuit (PLC) technology. There is known a WSS (see, for example, Patent Document 1) in which a light deflection unit is mainly configured by a liquid crystal switch. In this WSS, each input port and output port corresponds to one AWG. Since the signal light output to the space from the AWG connected to the input port has the same phase plane that differs for each wavelength, the propagation direction differs for each wavelength. Therefore, the AWG acts as a spectroscope having angular dispersion characteristics with different emission angles for each wavelength. Therefore, the signal light that has propagated through the AWG is collected by the lens at different positions for each wavelength in the x-axis direction. A liquid crystal switch is arranged at this focal position, and the liquid crystal switch is operated so as to be deflected at an arbitrary angle. With this spatial deflection function, the signal light can be optically coupled to an arbitrary multiplexing AWG via a lens. As described above, the entire optical system functions as a wavelength selective switch (WSS). At this time, assuming that the number of multiplexing AWGs is N, 1 × N-WSS having a function of one input and N outputs is obtained. This configuration integrates the input / output unit and the wavelength multiplexing / demultiplexing unit among the input / output unit, the wavelength multiplexing / demultiplexing unit, and the optical deflecting unit, which are the main functional units in WSS, to align the optical element. It has the feature that the mounting load derived from is reduced. Furthermore, AWG can be diffracted at a higher order than bulk diffraction gratings, so it has high dispersion and can use a lens with a shorter focal length. it can.

Japanese Patent No. 4949355 (paragraphs 0003 to 0005, 0016 to 0019, FIGS. 1 and 3) Japanese Patent No. 4960294 (paragraphs 0029 to 0030,0056, FIGS. 4, 5, and 8)

  However, in the WSS in which the wavelength multiplexing / demultiplexing unit is configured by a bulk diffraction grating that is a wavelength dispersion element, it is difficult to reduce the size of the optical system due to the limit of the dispersibility of the bulk diffraction grating. Angular dispersion (diffractive angle change per unit wavelength) of the bulk diffraction grating can be expressed by the following equation (1).

  In the formula (1), N is the number of grooves per mm, m is the diffraction order, and θ is the diffraction angle. As an example of a general diffraction grating, when the values of N = 940, m = 1, and θ = π / 4 are substituted, the angular dispersion is calculated to be about 1.3 rad / um. On the other hand, the angular dispersion of AWG that can be used as the wavelength multiplexing / demultiplexing unit of WSS can be expressed by the following equation (2).

  In equation (2), N is the effective refractive index, ΔL is the optical path length difference between adjacent exit points, and p is the pitch between adjacent exit points. As an example of a general AWG, when n = 1.5, ΔL = 200 μm, p = 10 μm, and λ = 1.55 μm are substituted, the angular dispersion is calculated to be about 19 rad / um, which is 10 times that of a bulk diffraction grating. Have dispersibility. There is a clear difference in dispersibility between commonly used bulk diffraction gratings and AWGs. In order to generate such a large difference, it is important to design p to be small while increasing ΔL in equation (2). In AWG, by controlling the waveguide layout, it is possible to set ΔL to be very large while maintaining p as it is, and therefore, it is possible to generate an orderly angular dispersion. On the other hand, in a bulk type diffraction grating, ΔL and p can be produced only with the same digit regardless of whether the production method is etching or replica. This is the reason why the angular dispersion cannot be set large with the current technology, and it is difficult to shorten the optical system as long as a bulk diffraction grating is used.

  As for WSS in which an input / output unit and a wavelength multiplexing / demultiplexing unit are integrated in an AWG, a small optical system can be realized as described above, but there is a problem that it is difficult to increase the number of input / output ports.

  FIG. 2 shows a conceptual diagram of a wavelength selective switch in which an input / output unit and a wavelength multiplexing / demultiplexing unit are integrated in an AWG, and the operation thereof will be described. In the following, in the wavelength selective switch, the direction of wavelength demultiplexing by the AWG is the x axis, the substrate thickness direction of the AWG 201 which is the direction orthogonal to the x axis is the y axis, and the traveling direction when the optical signal is output from the AWG. It is defined as the z axis. In this method, since one AWG corresponds to one input / output fiber in FIG. 1, the number of input / output ports is equal to the number of mounted AWGs.

  In the following, for the sake of explanation, the number of AWGs mounted on one PLC chip is three and two PLC chips are arranged side by side in the y-axis direction. The number of AWGs is not limited to this description.

  Since the AWG itself has a wavelength demultiplexing function, the signal light emitted from the AWG 201 to the space is demultiplexed in the x-axis direction and is condensed on the light deflection element 203 via the lens 202. In the optical deflection element 203, the light deflected at a desired angle with respect to the x axis or the y axis can be switched by being coupled to one AWG among the eight AWGs. At this time, since the AWG requires a certain chip area, the number of AWGs that can be mounted on one PLC is limited by the deflection capability of the optical deflection element 203, and usually only about 3 to 5 can be arranged. In order to arrange AWGs in the y-axis direction, either means of precisely aligning the PLC chips one by one or forming a laminated waveguide is adopted, but this is technically difficult and often Only about 4-5 layers can be realized.

  Therefore, in the WSS in which the input / output unit and the wavelength multiplexing / demultiplexing unit are integrated in the AWG, in order to increase the number of input / output ports, is it necessary to dramatically improve the basic characteristics of the optical deflecting element 203 or to greatly increase the mounting load? It is difficult to increase the number of ports.

  Further, it is conceivable that the wavelength multiplexing / demultiplexing unit of the WSS is configured by a VIPA (Virtually Imaged Phased Array) which is a highly dispersive element like the AWG. VIPA includes opposing parallel flat glass plates having mirror surfaces with different reflectivities, the reflectivity of one mirror surface is approximately 100%, and the reflectivity of the other mirror surface is lower than 100% (for example, 95 to 98%). By doing so, the incident optical signal is multiple-reflected between mirror surfaces, and is transmitted through a mirror surface having a reflectivity lower than 100% to be emitted. VIPA is known as an element that demultiplexes an input wavelength-multiplexed optical signal for each wavelength.

  The angular dispersion of VIPA can also be expressed by the above equation (2), which is the same value as the angular dispersion of AWG. Therefore, there is a clear difference in dispersion power between commonly used bulk diffraction gratings and VIPA. Exists. In order to generate such a large difference, as in AWG, VIPA can also set ΔL to be very large while controlling pitting angle and substrate thickness by controlling the installation angle and substrate thickness. Can be generated.

  Hereinafter, a case where the wavelength multiplexing / demultiplexing part of the WSS is configured with a VIPA including opposing parallel flat glass plates having mirror surfaces with different reflectivities will be considered.

  FIG. 3 shows a conceptual diagram of WSS using VIPA. The WSS shown in FIG. 3 includes an input / output fiber array 301 that constitutes an input / output unit, a microlens array 302, a VIPA 303 that constitutes a wavelength multiplexing / demultiplexing unit, a lens 304, and an optical deflection element 305 that constitutes an optical deflection unit. It is a structure arranged in order. Of the two opposing glass plates constituting the VIPA 303, the glass plate on the input / output fiber array 301 side has a mirror surface with a reflectance of almost 100%, and the glass plate on the lens 304 side has a reflectance of more than 100%. A low mirror surface is formed. The basic operation principle of the WSS shown in FIG. 3 is the same as the operation principle of the WSS using the AWG of FIG.

  The WSS shown in FIG. 3 can be expected to shorten the optical system in the same manner as the AWG by arranging a plurality of input / output fibers in the y-axis direction. On the other hand, the mounting load for laying out a plurality of wavelength multiplexing / demultiplexing portions with high accuracy by VIPA increases. Another problem unique to VIPA is that the intensity profile of the outgoing beam is asymmetric with respect to the x-axis. When the intensity distribution is asymmetric with respect to the x-axis, the intensity of the outgoing beam from the VIPA is collected through the lens 304, reflected by the light deflection element 305, and returned to parallel light by the lens 304 again. Since mismatch with the profile occurs, a fundamental coupling loss occurs. The light emitted from the input / output fiber array 301 is adjusted in NA by the microlens array 302 and enters the VIPA 303, but the beam diverges while propagating in space. Therefore, the beam diameter after being reflected a plurality of times by the VIPA 303 is larger than the beam diameter immediately after entering the VIPA 303 with respect to both the x-axis and the y-axis. For this reason, the mismatch of the intensity profile described above also occurs with respect to the y-axis. Therefore, when VIPA is used, it is an exponential intensity profile due to multiple reflections in the x-axis direction, and further, the beam diameter is increased by the divergence of a beam propagated through a long distance space due to multiple reflections. Loss becomes a problem.

  As described above, in the WSS proposed so far, the characteristics of the wavelength selective switch greatly depend on the characteristics of the wavelength multiplexing / demultiplexing unit. More specifically, the configuration is specialized for any of the merits such as downsizing, multiple ports, and low loss of the optical system, and it is difficult to realize a configuration that simultaneously realizes the above merits.

  The present invention has been made in view of such problems, and an object of the present invention is to provide a wavelength selective switch that can easily achieve downsizing and multi-port. Another object of the present invention is to provide a wavelength selective switch that can control the intensity profile of a beam emitted from a wavelength dispersion element into space.

In order to achieve such an object, the present invention provides an M × N wavelength selective switch, wherein at least M (M is an integer of 1 or more) input ports, at least N A number of output ports (N is an integer equal to or greater than 1) and a spectral means for wavelength-separating wavelength-multiplexed optical signals emitted from the input ports; Optical means, spatial deflecting means for changing the traveling direction of each of the optical signals collected by the condensing means, and optical signals whose traveling direction has been changed by the spatial deflecting means are combined and the N and a multiplexing means for emitting from one of output ports, each of said spectral means and said combining means, formed on the waveguide substrate, in the longitudinal direction of the pair of reflectors of the optical waveguide said optical guide Light vertically above and below the waveguide A pair of reflectors having a reflectivity of opposed asymmetrical at a distance of the order of the wavelength of No., the spectroscopic means, multiple reflections between optical signals input from said input port of said pair of reflecting mirrors While propagating in the longitudinal direction of the optical waveguide, it is emitted to the outside of the waveguide substrate at a plurality of reflection points at a predetermined refractive index angle determined by the wavelength of the optical signal. The optical signal is transmitted through the reflecting mirror and propagated in the longitudinal direction of the optical waveguide while being multiplexed and reflected between the pair of reflecting mirrors. And an M-input N-output function characterized in that the output is output from the output port to the outside .

The invention according to claim 2 is the M × N wavelength selective switch according to claim 1, wherein the optical waveguide includes a gain medium, and the waveguide substrate is an electrode for injecting energy into the gain medium from the outside. The optical signal is configured to amplify the light intensity when the optical signal is multiple-reflected between the reflecting mirrors.

A third aspect of the present invention is the M × N wavelength selective switch according to the first or second aspect , wherein the traveling direction is changed by the intensity distribution of the optical signal emitted from the spectroscopic means and the spatial deflecting means. It is characterized in that the intensity distribution of the optical signal transmitted through the condensing means and emitted to the combining means coincides.

A fourth aspect of the present invention is the M × N wavelength selective switch according to any one of the first to third aspects, wherein the light intensity distribution of the optical signal emitted from the spectroscopic means and the spatial deflection means are advanced. The light intensity distribution of the optical signal that changes direction, passes through the condensing means, and is emitted to the combining means includes a normal line including a point where the optical signal emitted from the spectroscopic means in the spatial deflecting means is condensed. It is characterized by being configured so as to be symmetrical and coincide with each other.

  A fifth aspect of the present invention is the M × N wavelength selective switch according to any one of the first to fourth aspects, wherein the intensity distribution of the optical signal emitted from the spectroscopic means is in the longitudinal direction of the optical waveguide. It is configured to have a Gaussian function.

A sixth aspect of the present invention is the M × N wavelength selective switch according to any one of the first to fifth aspects, wherein the spectroscopic means and the multiplexing means are formed on the same waveguide substrate. have a longitudinal separation groove of the optical waveguide between the spectroscopic unit and the multiplexing means and said.

  A seventh aspect of the present invention is the M × N wavelength selective switch according to any one of the first to sixth aspects, wherein M is 1.

  As described above, according to the present invention, two reflecting mirrors having asymmetric reflectance in the vertical direction perpendicular to the longitudinal direction of the optical waveguide are provided, and the distance between the reflecting mirrors is designed to be about the wavelength. A wavelength selective switch (WSS) having a configuration using a dispersion element having high angular dispersion, at least one lens, and an optical deflection element, and has a short focal length by using highly dispersed wavelength demultiplexing means. A lens can be introduced, and the device can be easily downsized. In addition, since a plurality of dispersive elements are monolithically integrated on a semiconductor substrate and manufactured collectively, the number of ports can be easily increased without increasing the mounting load. Furthermore, by adjusting the intensity profile of the beam emitted from the dispersive element to the space, it is possible to avoid an increase in the principle loss, which was a problem of the dispersive element due to multiple reflections, and to realize a low-loss wavelength selective switch. is there.

It is a conceptual diagram of a wavelength selective switch, (a) is a side view, (b) is a top view. It is a conceptual diagram of the wavelength selective switch using AWG, (a) is a side view, (b) is a top view. It is a conceptual diagram of the wavelength selective switch which comprised the wavelength multiplexing / demultiplexing part using VIPA, (a) is a side view, (b) is a top view. It is a conceptual diagram of the wavelength selective switch concerning embodiment of this invention, (a) is a side view, (b) is a top view. It is a schematic block diagram of the waveguide board | substrate which comprises the wavelength multiplexing / demultiplexing part of the wavelength selective switch concerning embodiment of this invention, (a) is sectional drawing, (b) is a top view. It is a conceptual diagram of the wavelength selective switch concerning embodiment of this invention, (a) is a side view, (b) is a top view. It is a schematic block diagram of the waveguide board | substrate which comprises the wavelength multiplexing / demultiplexing part of the wavelength selective switch concerning embodiment of this invention. It is a figure for demonstrating the generation | occurrence | production factor of the insertion loss in the wavelength selective switch concerning embodiment of this invention. It is a schematic block diagram of the waveguide board | substrate which comprises the wavelength multiplexing / demultiplexing part of the wavelength selective switch concerning embodiment of this invention. 4A and 4B are diagrams for explaining a principle loss reduction in the wavelength selective switch according to the embodiment of the present invention, where FIG. 5A is a top hat type profile, and FIG. 5B is a Gaussian type profile. It is. It is a schematic block diagram of the waveguide board | substrate which comprises the wavelength multiplexing / demultiplexing part of the wavelength selective switch concerning embodiment of this invention, (a) is sectional drawing, (b) is a top view.

  The present invention provides at least M (M is an integer greater than or equal to 1) input ports, at least N (N is an integer greater than or equal to 1) output ports, and spectral means for wavelength-separating wavelength multiplexed optical signals emitted from the input ports. A condensing means for condensing each of the optical signals separated by wavelength by the spectroscopic means, a spatial deflecting means for changing the traveling direction for each of the optical signals collected by the condensing means, and a spatial deflecting means Therefore, the optical signal can be implemented as an M × N wavelength selective switch including a multiplexing unit that multiplexes each of the optical signals whose traveling directions are changed and outputs them from any of the output ports. When the present invention is implemented with M = 1 and N = 1, the 1 × 1 wavelength selective switch is equivalent to a device (for example, a wavelength blocker) having a function capable of modulating the intensity of an optical signal.

  In one embodiment, the spectroscopic means is formed on the waveguide substrate, and includes a pair of upper and lower reflecting mirrors having asymmetric reflectance in the vertical direction perpendicular to the longitudinal direction of the optical waveguide, It has a structure of a lambda vertical resonator whose interval is adjusted to about the wavelength of the optical signal. The spectroscopic means formed on the waveguide substrate is configured such that an optical signal input from the optical input port repeats a plurality of reflections geometrically optically between a pair of reflecting mirrors in a longitudinal direction of the optical waveguide (multiple reflection). The light propagates and is emitted to the outside of the optical waveguide at a predetermined refractive index angle determined by the wavelength of the optical signal according to Snell's law at each of the plurality of reflection points.

Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS.
In the following examples, the substrate material of the wavelength demultiplexing section and wavelength multiplexing section (wavelength multiplexing / demultiplexing section) is GaAs, the distributed Bragg reflector (DBR) is a GaAlAs / GaAs semiconductor multilayer mirror, and the active region Although (gain medium) is illustrated as a GaInAs / GaAs multiple quantum well, it is by no means limited to this. Any structure may be used as long as the high refractive index film and the low refractive index film can be alternately stacked with a wavelength thickness of (1/4). Furthermore, any structure may be used as long as it is made of a semiconductor material capable of optical amplification by injecting current from outside. For example, it is needless to say that the substrate material is applicable to InP, DBR to an InGaAsP / InP semiconductor multilayer mirror, and the active region to an InGaAsP / InP multiple quantum well. Further, although the number of wavelength multiplexing is 2 in the first embodiment and 3 in the second embodiment, it goes without saying that the present invention is not limited to this.

(First embodiment)
The first embodiment will be described with reference to FIGS. Hereinafter, in the wavelength selective switch of the present invention, the direction parallel to the substrate surface of the waveguide substrate 403 on which the wavelength multiplexing / demultiplexing means is formed and the wavelength multiplexing / demultiplexing means performs wavelength demultiplexing is defined as the x axis. A direction parallel to 403 and perpendicular to the x-axis is defined as a y-axis, and a thickness direction of the waveguide substrate 403 is defined as a z-axis.

  FIG. 4 is a configuration diagram of a wavelength selective switch (WSS) according to the present embodiment. The WSS of this embodiment mainly includes an input optical fiber array 401, a microlens array 402, a waveguide substrate 403, a lens 404, and an optical deflection element 405. In the following, for the sake of explanation, the number of input / output fiber arrays 401 is five, the center fiber is set as the input port, and the fiber arranged at the bottom is set as the output port. Is not limited to this description. Furthermore, the number of lenses used is not limited as long as the same function is realized. In addition, the chief ray through which the optical signal emitted from the input fiber passes to the optical deflecting element 405 is thick, and the chief ray until the optical signal reflected by the optical deflecting element 405 is coupled to the output fiber is thick. It is represented by a broken line.

  FIG. 5 shows details of the waveguide substrate 403. According to FIG. 5, the waveguide substrate 403 is provided with a waveguide input port 501 for coupling the signal light emitted from the fiber array 401 and the microlens array 402 to the waveguide substrate 403 in the vertical direction of the substrate. It has an upper reflecting mirror 502, a lower reflecting mirror 503, and a waveguiding layer 504. The upper reflecting mirror 502, the lower reflecting mirror 503, and the waveguide layer 504 constitute wavelength multiplexing / demultiplexing means. The signal light coupled from the waveguide input port 501 to the substrate repeats multiple reflections geometrically and optically between the upper reflecting mirror 504 and the lower reflecting mirror 503. The distance between the reflecting mirrors in the vertical direction is about the wavelength, that is, about 1 μm. At this time, since the reflectivity of the upper reflector is designed to be lower than the reflectivity of the lower reflector, when propagating through the waveguide in a zigzag manner, a part of the light from the reflector having a low reflectivity enters the space. Emitted. At the time of emission, since different refraction angles are provided for each wavelength in accordance with Snell's law, the light is emitted to space at different angles for each wavelength of the input WDM optical signal. The optical signal emitted from the wavelength demultiplexing function unit of the waveguide substrate is demultiplexed for each wavelength as a result of multibeam interference in space. That is, the waveguide substrate 403 in this configuration operates as a wavelength demultiplexing element. Therefore, the function units configured on the waveguide substrate 403 are combined to form the wavelength demultiplexing function unit 500. In this embodiment, since the total number of input / output fiber arrays 401 is described as five, each wavelength demultiplexing function unit is arranged in a line in the y-axis direction as five corresponding to this number. Yes. Each wavelength demultiplexing function unit is optically independent. Therefore, in order to reduce crosstalk between adjacent wavelength demultiplexing function units, it is desirable to provide a light blocking layer between the wavelength demultiplexing function units. For example, although groove formation by etching is considered as one of the means, there is no problem in any form as long as the same function is realized.

At this time, the incident angle to the upper reflecting mirror is θ _i , the exit angle to the space is θ, the equivalent refractive index of the waveguide constituting the demultiplexing angle disperser is n _wg , and the refractive index of _air is n _air (= 1) ), N _air × sin θ = n _wg × sin θ _i is obtained according to Snell's law, and when the cutoff wavelength of the waveguide is λ _c and the wavelength used is λ, the refraction angle is expressed by the following relational expression: expressed.

Therefore, in this wavelength demultiplexing device, the angle θ at which the designed cutoff wavelength λ _c travels can be regarded as the basic traveling direction of the signal light, and the light with the wavelength λ _c behaves to travel at this angle. .

  The wavelength selective switch in this embodiment operates as follows. The optical signal emitted from the input / output fiber array 401 to the space is formed on the waveguide substrate 403 by adjusting the NA by the microlens array 402 whose focal length and arrangement position are adjusted so as to propagate through the space as collimated light. The wavelength demultiplexing function unit 500-3 is coupled. Since the waveguide substrate has a wavelength demultiplexing function, the optical signal coupled to the wavelength demultiplexing function unit 500-3 is wavelength demultiplexed in the x-axis direction, and further on the optical deflection element 405 via the lens 404. It is focused on. The optical signals of the respective wavelengths collected on the optical deflecting element 405 are adjusted in angle so as to be optimally coupled to the output fiber in the optical deflecting element 405 and enter the lens 404 again. Since the lens 404 has a function of converting the incident beam into an angle proportional to the height of the incident beam in the y-axis direction, the optical signal whose angle is adjusted by the light deflecting element 405 is demultiplexed by the condenser lens 404. The position and angle are adjusted so as to be parallel to the direction of the optical signal from the functional unit 500-3 toward the condenser lens 404, and the optical signal is coupled to the wavelength demultiplexing functional unit 500-5 and multiplexed. Then, the light can be switched by being emitted from the waveguide input port in the wavelength demultiplexing function unit 500-5 to the space and coupled to a desired output fiber in the input / output fiber array 401 via the microlens array 402. .

  The waveguide substrate 403 in this embodiment has a high angular dispersion based on the operating principle as described above, so that the entire optical system can be shortened, resulting in the merit of downsizing the entire device. Have. In addition, a plurality of wavelength demultiplexing function units are collectively manufactured on the same substrate with high accuracy by a photolithography process. Since the number of simultaneous wavelength demultiplexing function units is limited only by the size of the substrate, 20 or more wavelength demultiplexing function units can be easily arranged. This eliminates the need for individual alignment for each PLC chip as seen in a WSS that uses a plurality of arrayed waveguide diffraction gratings (AWGs). Furthermore, in the form using AWG, it is indispensable to realize stack mounting and laminated waveguides to increase the number of ports, and it was difficult to realize a very large number of ports. The number of ports can be easily realized because the parts can be manufactured in a lump. As described above, it is possible to provide a WSS that simultaneously realizes the two advantages of downsizing and multi-porting. In addition, since each wavelength demultiplexing function unit is optically independent, it is possible to suppress divergence in the y-axis direction in the waveguide, and the loss is lower than that of a conventional WSS using VIPA. WSS can be constructed.

  Furthermore, in this embodiment, by applying a two-dimensional optical deflection element that can be freely deflected in either the x-axis direction or the y-axis direction to the optical deflection element 405, it is possible to further increase the number of ports. A configuration example in this case is shown in FIG. In FIG. 6, the number of output ports is doubled by using two sets of the input / output fiber array 601, the microlens array 602, and the waveguide substrate 602. In FIG. 6, which is an explanatory diagram, two waveguide substrates 603-1 and 603-2 are separately arranged, but both may be realized by a single waveguide substrate. In this case, as shown in FIG. 7, the wavelength demultiplexing function unit 701 corresponding to the waveguide substrate 603-1 and the wavelength demultiplexing function unit 702 corresponding to the waveguide substrate 603-2 are two-dimensionally arranged on one substrate. By disposing it above, it is possible to further reduce the mounting load without requiring additional member preparation and alignment in mounting.

(Second embodiment)
A second embodiment will be described with reference to FIGS. In the first embodiment described above, it is described that the waveguide substrate 403 has a wavelength demultiplexing function as a result of multi-beam interference in space due to multiple reflection of optical signals propagating in the substrate. A configuration example of the wavelength selective switch using the function is shown. Further, in the first embodiment, each wavelength demultiplexing function unit is optically independent and can suppress beam spread in the y-axis direction, so that it is low compared with the conventional WSS using VIPA. A lossy WSS can be constructed. However, it is known that a loss is generated in principle by the intensity profile in the x-axis direction of the beam emitted from the wavelength demultiplexing function unit. FIG. 8 is a diagram illustrating this phenomenon.

  In FIG. 8, after entering the wavelength demultiplexing function unit, the locus of the light output to the space by the first reflection is indicated by a bold line, and the locus of the light reflected last and output to the space is indicated by a broken line. ing. As the intensity of light, the thick line locus is the strongest and the broken line locus is the weakest. Further, from the viewpoint of the propagation distance in the wavelength demultiplexing function unit, the thick line is the shortest, and the broken line is emitted to the space after propagating the longest distance. In this way, the light source group given the optical path length difference interferes and wavelength demultiplexes, but in order to multiplex the light in the same wavelength demultiplexing function unit, the optical path length difference of each light The light path must be set to cancel. That is, when the light that follows the thick line is coupled to the wavelength demultiplexing function unit again, it is propagated through the longest distance in the wavelength demultiplexing function unit by coupling to the emission point position of the broken line, and the light that follows the broken line is again wavelength demultiplexed. When coupling to the functional unit, it is necessary to propagate the longest distance in the wavelength demultiplexing functional unit by coupling to the emission point position of the thick line. In such a case, both phases are matched, and wavelength multiplexing is possible. However, since the intensity profile at the time of emission and the intensity profile after reflection do not match with each other as it is, a fundamental loss due to this occurs. Although the WSS in the first embodiment has a configuration capable of sufficiently reducing the loss even if this principle loss is included, a node very sensitive to the loss may be required to have a loss as low as possible.

  FIG. 9 shows the configuration of the wavelength demultiplexing function unit for avoiding the above-described loss. According to FIG. 9, the waveguide substrate 403 in the first embodiment includes a waveguide input port 901 for coupling the signal light emitted from the fiber array 401 and the microlens array 402 to the waveguide substrate 403, and a waveguide. An upper reflecting mirror 902, a lower reflecting mirror 903, a waveguiding layer and an active region 904, and electrodes 905 and 906 formed in a direction perpendicular to the waveguide substrate are provided. In the waveguide substrate 403, a light blocking layer (for example, a groove formed by etching) may be provided between the wavelength demultiplexing function units. The basic wavelength demultiplexing operation is the same as that of the first embodiment, but an active region is provided in the waveguide layer 904, electrodes 905 and 906 are provided, and current can be injected into the substrate. There are some differences.

  When the waveguide layer 904 has an active region and has a structure of a semiconductor optical amplifier, signal light can be amplified by injecting a current. In this embodiment, current is injected by the electrodes 905 and 906 and optical amplification is performed, so that the fundamental coupling loss in the multi-reflection type demultiplexing element such as the present invention or the VIPA described above can be compensated without additional members. It is possible to simultaneously realize the new merit of low loss while maintaining the merit of downsizing and multi-port in the first embodiment.

(Third embodiment)
A third embodiment will be described with reference to FIGS. In the second embodiment described above, the example of WSS that compensates the principle loss due to mismatch of intensity profiles by the optical amplification effect of the optical semiconductor has been described. On the other hand, if the intensity profile of the beam emitted from the wavelength demultiplexing function unit and the intensity profile reflected by the optical deflection element and recombined with the wavelength demultiplexing function unit can be matched, the principle loss is reduced. it can. FIG. 10 illustrates this.

  In FIG. 10, only the xz plane of WSS is shown for simplicity of explanation, but in this description, as shown in FIG. 4 or FIG. It is noted that there are a plurality of wavelength demultiplexing function units that are present in the direction, and a wavelength demultiplexing function unit that is recombined after being reflected by the optical deflection element. The normal line of the reflection surface of the optical deflection element that extends from the position where the beam emitted from the wavelength demultiplexing function unit is condensed on the optical deflection element will be described as line A. Although the line A is described as a line passing through the center of the lens, the line A does not necessarily pass through the center of the lens, and may not be a normal to the reflection surface of the light deflection element. As will be described below, it is important that a condition that the intensity profiles in the wavelength demultiplexing function unit match is satisfied. As long as this condition is satisfied, no matter how optically the line A is optically designed, there is no problem. Absent. When the intensity profile of the output beam from the wavelength demultiplexing function section is symmetrical with respect to the line A with respect to the wavelength demultiplexing axis direction, for example, a top hat type, or a Gaussian type in which the maximum intensity appears on the line A As for, the intensity of the thick line and the intensity of the broken line are equal, and when the light emitted from the wavelength demultiplexing function unit is recombined, it is coupled with maximum efficiency. Accordingly, the principle loss can be reduced by matching the beam profile emitted from the wavelength demultiplexing function unit with the beam profile of the light reflected by the optical deflection element.

  The configuration of the wavelength demultiplexing function unit for avoiding the above-described loss is shown in FIG. According to FIG. 11, the waveguide substrate 403 according to the third embodiment includes a waveguide input port 1101 for coupling the signal light emitted from the fiber array 401 and the microlens array 402 to the space, and a waveguide. An upper reflecting mirror 1102, a lower reflecting mirror 1103, a waveguiding layer and active region 1104 formed in the vertical direction of the waveguide substrate 403, an electrode array 1105, an electrode separation groove 1106, and an electrode 1107 are provided. The basic wavelength demultiplexing operation is the same as in the first and second embodiments, but an electrode array 1105 in which an active region is provided in the waveguide layer 1104 and a plurality of electrodes are arranged, and an electrode array The difference is that an electrode separation groove 1106 and an electrode 1107 are provided between the electrodes in 1105.

  This embodiment is characterized in that current injection is individually performed for each of the electrodes 1105. When such a configuration is adopted, different gains are received depending on the positions in the wavelength demultiplexing function unit, and the intensity profile can be freely adjusted. For example, in FIG. 11, when the current injection amount is adjusted so that the electrode disposed farther from the waveguide input port 1101 has a larger gain, the gain of the portion that is originally emitted into the space with high intensity is low, and the opposite In addition, since the gain of the portion that is emitted into the space with low intensity is high, the intensity of all the light emitted from the wavelength demultiplexing function unit 1100 can be made equal. This is equivalent to shaping a beam having a so-called top-hat intensity distribution. Therefore, the intensity profile of the beam emitted from the wavelength demultiplexing function unit and the intensity profile of the beam that is recombined after being reflected by the optical deflecting element match, and the principle loss can be completely eliminated. It becomes.

  Such an effect is not obtained only by shaping a beam symmetrical with respect to the line A in the wavelength demultiplexing axis direction. For example, the beam emitted from the wavelength demultiplexing function unit 1103-3 is an exponential function. Even if it is a distorted profile, it is possible to avoid the principle loss by adjusting the current injection amount so that another wavelength demultiplexing function unit 1103-5 to which this beam is combined matches the distorted intensity profile. is there. It is only necessary that the intensity profile of the beam is adjusted to a specific shape, and that the intensity profile is the same in the wavelength demultiplexing function unit where the beams reflected by the light deflecting elements are combined. It is an important point. In this embodiment, in addition to the fact that the loss due to mismatch of intensity profiles can be improved in principle as described above, the optical amplification function of the semiconductor optical amplifier is provided to further reduce the loss of the device. Can be achieved. Of course, this merit can coexist with the merit of downsizing and multi-porting as in the first and second embodiments, and it is possible to provide a WSS that simultaneously realizes these merit.

DESCRIPTION OF SYMBOLS 100 Input / output part 101,301,401,601 Input / output fiber array 102,302,402,602 Micro lens array 103 Condensing lens 104,106,202,304,404,604 Lens 105 Diffraction grating 107,203,305, 405, 605 Optical deflection element 201 Array waveguide grating 303, 403, 603 VIPA
500, 1100 Wavelength demultiplexing function section 501,901,1101 Waveguide input port 502,902,1102 Upper reflecting mirror 503,903,1103 Lower reflecting mirror 504 Waveguide layer 701,702 Wavelength demultiplexing function section array 904,1104 Wave layer and active region 905, 906, 1107 Electrode 1105 Electrode array

Claims (7)

At least M (M is an integer greater than or equal to 1) input ports, at least N (N is an integer greater than or equal to 1) output ports, and spectral means for wavelength-separating wavelength multiplexed optical signals emitted from the input ports; Condensing means for condensing the optical signals spectrally separated by each wavelength by the means, spatial deflecting means for changing the traveling direction for each of the optical signals collected by the condensing means, and the spatial deflecting means And combining means for combining each of the optical signals whose traveling directions have changed by the above and outputting them from any of the N output ports,
The spectroscopic means and the multiplexing means are each a pair of reflecting mirrors formed in the waveguide substrate in the longitudinal direction of the optical waveguide, and are opposed to each other with an interval of about the wavelength of the optical signal vertically above and below the optical waveguide. and a pair of reflectors having a reflectivity of asymmetrical, the spectroscopic means, the optical signal inputted from the input port propagates in the longitudinal direction of the optical waveguide with multiple reflections between said pair of reflecting mirrors , At a plurality of reflection points, the light is emitted to the outside of the waveguide substrate at a predetermined refractive index angle determined by the wavelength of the optical signal, thereby performing spectroscopy with a different refraction angle for each wavelength, and the multiplexing means Each of the optical signals that are transmitted through the reflecting mirror propagates in the longitudinal direction of the optical waveguide while being multi-reflected between the pair of reflecting mirrors, and are combined and output from the output port to the outside. M input N output characterized by M × N wavelength selective switch, characterized in that a function. The optical waveguide includes a gain medium, and the waveguide substrate includes an electrode for injecting energy from the outside into the gain medium, and is configured to amplify the light intensity when the optical signal is multiple-reflected between the reflecting mirrors. The M × N wavelength selective switch according to claim 1, wherein: And intensity distribution of the light signals emitted from said spectral means, is configured such that the intensity distribution in the traveling direction by said spatial deflection means changes transmitted through the condensing means an optical signal to be emitted to the combining means is coincident The M × N wavelength selective switch according to claim 1, wherein the M × N wavelength selective switch is provided. The light intensity distribution of the light signal emitted from the spectroscopic means and the light intensity distribution of the light signal transmitted through the condensing means and emitted to the combining means are changed by the spatial deflection means. 4. The M × according to claim 1, wherein the optical signal emitted from the spectroscopic means in the means is configured so as to be symmetrical and coincide with a normal line including a condensing point. 5. N wavelength selective switch.   5. The M × N wavelength according to claim 1, wherein the intensity distribution of the optical signal emitted from the spectroscopic means is configured to have a Gaussian function in the longitudinal direction of the optical waveguide. Select switch. Wherein the spectroscopic means and the combining means are formed on the same waveguide substrate, according to claim 1, characterized in that it comprises a longitudinal separating groove of the optical waveguide between the multiplexing means and the dispersing means The M × N wavelength selective switch according to any one of 1 to 5.   The M × N wavelength selective switch according to claim 1, wherein M is one.

Images