JP4885996B2 - Dispersion compensator - Google Patents

Description

  The present invention relates to a dispersion compensator used in optical fiber communication.

  With the explosive spread of the Internet, wavelength division multiplexing (WDM) communication is moving from a conventional point-to-point system to a ring-mesh system. This is because a ring mesh type system can flexibly cope with changes in communication demand between nodes by using a transparent wavelength selective switch or the like that processes an optical signal in an optical state. However, in a ring mesh type network, the dispersion value of the path dynamically changes as the optical path is switched. For this reason, adaptability is also required for dispersion compensation of optical communication paths. Conventional dispersion compensators are mainly of a type that collectively compensates for a plurality of channels. However, in a ring mesh type network using a wavelength selective switch, the distance of the path that passes is different for each wavelength. For this reason, there has been a demand for setting a different dispersion value for each WDM wavelength.

  As an adaptive dispersion compensation technique that meets such a demand, for example, a technique using a spectroscope and a mirror array (Patent Document 1), a technique using a waveguide (Patent Document 1, Non-Patent Document 1), a three-dimensional mirror, The thing by a spectroscopic element (nonpatent literature 2) etc. was proposed.

  In addition, a variable dispersion compensator with a smaller size and lower cost using an LCOS (Liquid Crystal on Silicon) type spatial phase modulator by combining an arrayed waveguide grating (AWG) and a spatial optical system is also proposed. Has been. The tunable dispersion compensator using AWG has a feature that a large dispersion value can be set by using a high order diffraction order from several tens to several thousand based on the flexible optical design of AWG. . If the relationship between the bandwidth of the communication channel and the AWG FSR (Free Spectral Range) design value is appropriately set, dispersion compensation can be performed for a plurality of channels at once. By combining an AWG with an LCOS type spatial phase modulation element, it was possible to set a more flexible dispersion compensation value.

  FIG. 20 is a diagram illustrating a configuration example of a tunable dispersion compensator according to the related art. FIG. 20A shows a configuration of a tunable dispersion compensator (TODC) in which one WDM communication channel and the AWG FSR value are matched. The TODC is composed of an AWG, a condensing lens, a spatial phase modulation element, and the like. The AWG FSR is configured to correspond to one WDM communication channel. According to the configuration of FIG. 20A, dispersion compensation can be performed collectively for 40 WDM communication channels from λ1 to λ40, for example.

  FIG. 20B is a diagram showing the light transmittance of the variable dispersion compensator having the configuration shown in FIG. It can be seen that on the optical frequency axis, signal processing is collectively performed for 40 communication channels that coincide with the cycle of the FSR. However, dispersion compensation cannot be performed independently for each channel.

  FIG. 21A is a diagram illustrating a configuration example of a dispersion compensator capable of performing dispersion compensation independently for each communication channel. This tunable dispersion compensator includes an AWG 34, a cylindrical lens 35, a condenser lens 36, an LCOS element 37, and the like. This configuration is a reflection-type variable dispersion compensator in which an optical signal is reflected by an LCOS element and a single AWG 34 serves as both demultiplexing and multiplexing of the optical signal. The optical signal group subjected to dispersion compensation is input to the input port 38 of the optical circulator 11. The dispersion-compensated optical signal group is output from the output port 39 of the optical circulator 11. A plurality of pixel groups arranged in the spectral axis direction (x-axis) of the AWG 34 on the LCOS element 37 are divided into six groups, and one WDM communication channel (ch1,... Ch6) is assigned to each group of pixel groups. (Non-patent Document 3). Dispersion compensation can be performed independently for each channel by giving a different phase distribution φ (x) to each group of pixels on the LCOS element. FIG. 21B shows group delay characteristics set independently for each channel.

  FIG. 22 is a diagram illustrating an example of setting the phase distribution for one WDM communication channel in the variable dispersion compensator having the configuration of FIG.

JP 2002-303805 (pages 5-7, FIG. 1 and FIG. 11) K. Takiguchi, K. Okamoto, and T. Goh, “Dispersion slope equalizer on planar Lightwave circuit for 40Gbit / s based WDM transmission,” Electron. Lett, 37 (24), p.1469-1470, 2001. National Institute of Information and Communications Technology, “Research and Development Report of FY 2004 Research and Development of Highly Functional Integrated Optical Switching Nodes to Realize Economical Optical Networks,” 2006 K. Seno, K. Suzuki, K. Watanabe et al., “Channel-by-channel tunable optical dispersion compensator consisting of arrayed-waveguide grating and liquid crystal on silicon” OWP4, Proceeding of OFC2008 http://surf.ap.seikei.ac.jp/fopt/Kougaku-Houkoku.pdf (Seikei University Engineering Research Report Vol.39 pp.37-41, 2002)

  However, the conventional variable dispersion compensator using the spatial phase modulation element has the following problems. As described above, according to the configuration of FIG. 20A, dispersion compensation could not be performed independently for each WDM communication channel. Further, according to the configuration of FIG. 21A, dispersion compensation can be performed independently for a plurality of WDM communication channels, but the required spatial phase modulation element becomes large. There was a problem. That is, in order to realize dispersion compensation for many channels, a spatial phase modulation element that is long in a direction corresponding to the spectral axis direction of the AWG is required.

  For example, in the configuration of FIG. 21A, dispersion compensation is performed for six communication channels having a bandwidth of 100 GHz. Here, when the number of communication channels increases, the size of the LCOS element in the spectral axis direction (x-axis direction) increases. More specifically, the number of LCOS elements corresponding to one WDM communication channel is 128, and the pixel arrangement pitch is 8 μm. At this time, the length of the LCOS element per channel in the spectral axis (x-axis) direction needs to be about 1000 μm. Therefore, the total length of the LCOS element exceeds 40 mm. If the size of the LCOS element in the direction of the spectral axis is to be shortened, there is a concern that characteristics such as a dispersion compensation amount and a channel transmission band are degraded. Therefore, depending on the required chromatic dispersion value and the number of channels, an LCOS element having a large number of pixels and a very long side is required.

  As is generally known, as the chip size increases, the mass productivity decreases and the cost increases. An increase in the size of the spatial phase modulation element is not preferable in terms of mass productivity and manufacturing cost. Therefore, in a multi-channel variable dispersion compensator using a spatial phase modulation element, it has been required to realize mass productivity and low cost.

  In addition, when a tunable dispersion compensator is configured using a spatial phase modulation element, there is a problem that light transmission characteristics are deteriorated at the optical frequency (wavelength) corresponding to the end of the AWG FSR. There is also a problem in using a pixel corresponding to the end of the FSR in the pixel group of the spatial phase modulation element.

  The present invention has been made in view of the above-described problems, and an object thereof is to realize a multi-channel variable dispersion compensator that realizes higher mass productivity and low cost. Furthermore, a multi-channel variable dispersion compensator with a more flat light transmission characteristic is realized by effectively using pixels of the spatial phase modulation element corresponding to the FSR. Maintenance of the tunable dispersion compensator is simplified and the cost is reduced.

In order to achieve such an object, the present invention provides a plurality of optical signal groups of N consecutive communication channels including a plurality of channels that are discretely selected at a predetermined channel interval. In a dispersion compensator for an interleaved communication system that performs processing by branching to a plurality of wavelength groups, the optical signal groups of the N communication channels are represented by b first wavelengths each having a maximum of a communication channels. a group branching filter of interleaved to demultiplexing the group, the a b number of dispersion compensator block corresponding to each of the b-number first wavelength groups, each dispersion compensator block, the number b A second group demultiplexing filter that further multiplexes and demultiplexes a corresponding one of the first wavelength groups into two sub-wavelength groups, the first of the b first wavelength groups. One of corresponding to the wavelength group, the N communication Against Yan'neru, a second group branching filter having continuous by transmitting light signals for each a number of channels adjacent disposed or block filter characteristics, the optical signal group in said sub-wavelength group Further, it is an optical multiplexing / demultiplexing means for demultiplexing, and when the bandwidth of one communication channel is B, it has an FSR of (2 × B × a), and each has an optical frequency difference corresponding to 1/2 FSR. An optical multiplexing / demultiplexing unit that includes two input / output ports and selects the input / output port corresponding to each of the two sub-wavelength groups , and a plurality of elements arranged in the demultiplexing axis direction of the optical multiplexing / demultiplexing unit a spatial phase modulator including an element, the plurality of element devices comprises a spatial phase modulation element is divided into a number of compartments, each corresponding to one communication channel, a and b are each other A ≧ 3, b The dispersion compensator is selected so as to satisfy the relationship of ≧ 2 and N ≦ a × b. Note that the interleaved group demultiplexing filter is also referred to as an interleaver. The second group demultiplexing filter corresponds to TFF.

  The invention according to claim 2 is the dispersion compensator according to claim 1, wherein the multiplexing / demultiplexing means is an arrayed waveguide diffraction grating (AWG), and the spatial phase modulation element is LCOS. To do.

  The invention described in claim 3 is the dispersion compensator according to claim 1, wherein a power of 2 is selected for b.

The invention according to claim 4 is the dispersion compensator according to claim 1, wherein in each of the spatial phase modulation elements, the distance of the demultiplexing axis is independent of each of the sections corresponding to each of the optical signals. And a phase defined by a function of second order or higher is set as a parameter.

A fifth aspect of the present invention is the dispersion compensator according to any one of the first to fourth aspects, wherein the spatial phase modulation element reflects an incident optical signal corresponding to the communication wavelength, and wave means, said optical demultiplexing means, characterized in that it is a reflective type configuration for multiplexing the dispersion compensated optical signal.

  The invention according to claim 6 is the dispersion compensator according to claim 5, further comprising a circulator for inputting / outputting optical signal groups of the N communication channels on an input side of the interleaved group demultiplexing filter. It is characterized by having.

  As described above, according to the present invention, a plurality of variable dispersion compensator blocks (hereinafter referred to as TODC blocks) using spatial phase modulation elements in which the size in the direction corresponding to the spectral axis of the AWG is kept short are combined, and more A multi-channel variable dispersion compensator that realizes good mass productivity and low cost can be realized. Maintenance and operation of the tunable dispersion compensator can also be simplified and reduced in cost. Furthermore, the type of TODC block can be reduced by setting the AWG FSR used in the TODC block to be equal to the entire bandwidth of the WDM communication channel group covered by the TODC block. Furthermore, by setting the FSR so as to have a predetermined relationship with the total bandwidth of the WDM communication channel group covered by the TODC block, the light transmission characteristics can be flattened. Furthermore, even when a group demultiplexing filter is used in the interleave type, the number of wavelength groups and the FSR can be appropriately selected to realize flattening of light transmission characteristics. Channels can also be efficiently assigned to cells on LCOS (spatial phase modulation elements) using an interleaver or in combination with a group demultiplexing filter.

1 is a block diagram illustrating an overall configuration of a tunable dispersion compensator according to Embodiment 1. FIG. FIG. 3 is a diagram illustrating light transmittance characteristics of the tunable dispersion compensator according to the first embodiment. 6 is a configuration diagram of a TODC block used in the variable dispersion compensator of Embodiment 2. FIG. It is a figure which shows the light transmittance of the TODC block which concerns on Example 2. FIG. FIG. 6 is a block diagram illustrating an overall configuration of a tunable dispersion compensator according to a second embodiment. 6 is a configuration diagram of a TODC block used in the variable dispersion compensator of Example 3. FIG. 6 is a configuration diagram of an LCOS element used in a TODC block used in a tunable dispersion compensator of Example 3. FIG. FIG. 9 is a block diagram illustrating an overall configuration of a tunable dispersion compensator according to a third embodiment. FIG. 10 is a configuration diagram of a TODC block used in the variable dispersion compensator of Example 4. FIG. 10 is a block diagram illustrating an overall configuration of a tunable dispersion compensator according to a fourth embodiment. FIG. 10 is a block diagram illustrating an overall configuration of a variable dispersion compensator according to a fifth embodiment. FIG. 10 is a diagram illustrating a more detailed configuration of a variable dispersion compensator according to a fifth embodiment. FIG. 10 is a conceptual diagram showing the channel number and LCOS cell assignment and the corresponding TFF filter characteristics in the variable dispersion compensator of Example 5 in association with each other. It is a figure explaining the channel allocation with respect to the LCOS cell in the variable dispersion compensator of Example 5. FIG. FIG. 10 is another diagram illustrating channel assignment for LCOS cells in the tunable dispersion compensator of Example 5. It is the figure which showed the channel allocation result with respect to the LCOS cell in the variable dispersion compensator of Example 5. FIG. FIG. 10 is a diagram illustrating a configuration of an AWG applicable to the tunable dispersion compensator according to the fifth embodiment. FIG. 10 is a diagram showing another AWG configuration applicable to the variable dispersion compensator of Example 5. FIG. 10 is a diagram showing the relationship between the total number of wavelengths and the number of LCOS elements applicable to the tunable dispersion compensator of Example 5. It is a figure which shows the structure of the variable dispersion compensator of the prior art using a spatial phase modulation element. It is a figure which shows the other structure of the variable dispersion compensator of the prior art using a spatial phase modulation element. It is a figure which shows the example of the phase setting in the variable dispersion compensator of the prior art using a spatial phase modulation element.

  In the present invention, a multi-channel variable dispersion compensator is configured by combining a plurality of variable dispersion compensator blocks (TODC blocks) using spatial phase modulation elements whose size in the direction corresponding to the spectral axis of the AWG is kept short. Has characteristics. Further, the AWG FSR used in the TODC block is set to be equal to the entire bandwidth of the WDM communication channel group covered by the TODC block. Furthermore, by setting the FSR so as to have a predetermined relationship with the total bandwidth of the WDM communication channel group covered by the TODC block, the light transmission characteristics can be flattened. Channels can also be allocated efficiently using an interleaver or in combination with a group demultiplexing filter. Hereinafter, the present invention will be described in detail with examples.

FIG. 1 is a block diagram showing an overall configuration of a tunable dispersion compensator according to Embodiment 1 of the present invention. The tunable dispersion compensator 100 is composed of a plurality of TODC blocks and a group demultiplexing filter. For example, an optical signal in which an optical signal group having 40 wavelengths from λ ₁ to λ ₄₀ is multiplexed is input to the input port In of the first group demultiplexing filter 101. The multiplexed optical signal is divided into four wavelength groups (G1, G2,..., Λ ₁ to λ ₁₀ , λ ₁₁ to λ ₂₀ , λ ₂₁ to λ ₃₀ and λ ₃₁ to λ ₄₀ by the first group demultiplexing filter 101. G3, G4) are group-demultiplexed. The optical signal group of each wavelength group output from the port 1, port 2, port 3, and port 4 of the wavelength group demultiplexing filter 101 is input to the TODC blocks 103a, 103b, 103c, and 103d through optical fibers or the like, respectively. . The optical signal groups of the wavelength groups G1 to G4 are compensated for dispersion by the TODC blocks 103a, 103b, 103c, and 103d, respectively, and are connected to the ports 1 to 4 of the second wavelength group demultiplexing filter 102. Optical signal group of the four wavelength groups after dispersion compensation (G1, G2, G3, G4 ) are multiplexed by the second wavelength group branching filter 102, 40 again having a central wavelength of from lambda ₁ to lambda ₄₀ Are multiplexed on the optical signal of the communication channel and output from the output port out.

  In the block diagram shown in FIG. 1, it is described that there are two wavelength group demultiplexing filters 101 and 102. However, when a reflection type configuration is used as the TODC block, a reflection plate is provided. An equivalent function can be realized by one wavelength group demultiplexing filter capable of generating waves and demultiplexing. The wavelength group demultiplexing filters 101 and 102 can be composed of, for example, a dielectric multilayer film.

  As each TODC block 103 of the tunable dispersion compensator of the present invention, for example, the same configuration as that of the tunable dispersion compensator shown in FIG. Therefore, the input and output of each TODC block in FIG. 1 correspond to the input port 38 and the output port 39 of the optical circulator 11 in FIG. In FIG. 23A, a variable chromatic dispersion compensator having a reflection type is shown, but it should be noted that it can also be realized by a transmission type configuration in which two AWGs are arranged.

  For the individual TODC blocks 103a, 103b, 103c, 103d, the configuration parameters of the AWG are set so that the center wavelength of the spectroscopic operation corresponds to the center wavelength of each band of G1-G4, and the predetermined number of communication channels in each band Each is set so that the linear dispersion value conforms to. The basic configuration of each TODC block can be exactly the same. As shown in FIG. 23A, the phase can be set independently for each channel by a plurality of pixels arranged in the AWG's spectral axis direction on the LCOS element included in each TODC block. . In the case of the configuration of FIG. 1, dispersion compensation can be performed independently for 10 WDM communication channels in one wavelength group band.

  Therefore, with the configuration shown in FIG. 1, the size corresponding to the spectral axis of each LCOS element can be shortened by suppressing the number of communication channels for performing dispersion compensation in each TODC block. That is, it is possible to realize a multi-channel variable dispersion compensator excellent in mass productivity and cost while suppressing an increase in size of the LCOS element. In particular, when the number of channels requiring dispersion compensation is very large, a smaller LCOS element is used compared to the case where one TODC block using only one LCOS element having a large side is used. It is cheaper to configure with a plurality of TODC blocks.

  More preferably, in each TODC block of the present embodiment, the AWG FSR can be set to be equal to the entire bandwidth of the communication channel included in one wavelength group. Specifically, the AWG FSR can be set to an optical frequency width corresponding to 10 channels of the WDM communication channel. For example, if one WDM communication channel width is 100 GHz, the FSR is set to 1000 GHz.

  By setting the AWG FSR and the total bandwidth of a plurality of communication channels included in one wavelength group to the same value, the configurations of the TODC blocks 103a, 103b, 103c, and 103d may be exactly the same. it can. That is, the tunable dispersion compensator 100 shown in FIG. 1 can be configured with only one type of TODC block having the same design specifications. Since the types of components of the tunable dispersion compensator 100 can be reduced, the manufacturing cost of the tunable dispersion compensator can be reduced. Furthermore, from the viewpoint of maintenance and operation of the tunable dispersion compensator, maintenance work can be simplified and cost can be reduced. In order to cope with the failure of the tunable dispersion compensator, it is not necessary to provide a plurality of types (for example, four types) of TODC blocks, and only one type of TODC block may be provided for maintenance. Note that maintenance and replacement can be simplified.

  FIG. 2 is a diagram showing an example of the light transmittance characteristic of the variable dispersion compensator of the present invention. In this example, the AWG FSR is set to the same value as the total bandwidth of the communication channels included in one wavelength group. The horizontal axis indicates the optical frequency, and the vertical axis indicates the light transmission characteristics as a dispersion compensator. The dispersion compensation characteristic is repeated for each FSR due to the circulatory property of the AWG. As for the light transmittance, band characteristics are repeated for each FSR. One FSR corresponds to the wavelengths of ten communication channels in one wavelength group (G1, G2, G3, G4).

  Although the variable dispersion compensator having the configuration shown in FIG. 1 has been described as an example in which 10 WDM communication channels are included in one wavelength group, the present invention is not limited to this. Similarly, the total number of channels of the system subject to dispersion compensation is 40, the number of wavelength groups is 4, and the number of corresponding TODC blocks is 4. However, these numbers are not limited at all.

  According to the variable chromatic dispersion compensator of the first embodiment described above, the multiplexed WDM optical signal is separated into a plurality of wavelength groups, and a TODC block corresponding to each wavelength group is provided, so that the spectral axis direction can be provided. LCOS elements with a short size can be used. A multi-channel variable dispersion compensator excellent in mass productivity and cost can be realized without requiring a large LCOS element. In addition, by setting the AWG FSR and the total bandwidth of the communication channels included in one wavelength group to the same value, it is possible to reduce the cost of the variable dispersion compensator, simplify the maintenance, and reduce the cost. it can.

  In the present embodiment, a tunable dispersion compensator is shown in which a TODC block having completely the same specification can be used and the light transmission characteristics are flattened. In the first embodiment, the basic configuration of each TODC block can be the same. However, unless the AWG FSR and the total bandwidth of the communication channels included in one wavelength group are set to the same value, it is necessary to prepare a dedicated TODC block corresponding to each wavelength group. That is, for each wavelength group, it is necessary to set the configuration parameters of the AWG included in the TODC block so that the center wavelength of the spectral operation of the AWG corresponds to the center wavelength of each band of G1 to G4. The AWG chip constituting the TODC block must be different for each TODC block.

When the AWG FSR and the total bandwidth of the communication channels included in one wavelength group are set to the same value, one type of TODC block can be used. However, as shown in FIG. 2, it is inevitable that the light transmittance decreases in the communication channel corresponding to both ends of the FSR. When the AWG FSR and the entire bandwidth of the communication channel included in one wavelength group are set to the same value, the transmittance of the dispersion compensator also shows a characteristic that repeats for each FSR. As shown in FIG. 2, in general, the transmittance decreases in the communication channels at both ends of one wavelength group. For example, in the communication channels corresponding to λ ₁ and λ ₁₀ , there is a problem that the light transmittance is reduced as compared with the communication channel in the center of one wavelength group. In this embodiment, this problem is further improved.

  FIG. 3 is a diagram illustrating a configuration of a TODC block used in the variable dispersion compensator according to the second embodiment. The TODC block 103 is used as a component of a tunable dispersion compensator according to a second embodiment which will be described later with reference to FIG. The basic configuration of the TODC block 103 shown in FIG. 3 is the same as the configuration of the variable dispersion compensator in the prior art shown in FIG. Further, as in the phase distribution shown in FIG. 24, it is defined by a function of second order or higher with the distance of the demultiplexing axis as a parameter. The AWG is different from the configuration of the prior art in that the AWG has two input / output waveguides connected to positions separated by 0.5 FSR on the boundary surface of the slab waveguide.

  Referring to FIG. 3, the TODC block includes a spatial phase modulation element 8 including an AWG 1, a cylindrical lens 6, a condenser lens 7, and an LCOS element. The AWG 1 includes a slab waveguide 3 and an arrayed waveguide 4. The multiplexed optical signal group is demultiplexed from one end of the arrayed waveguide 4 at an emission angle corresponding to the wavelength of the optical signal, and emitted from the end face A of the AWG substrate. On the LCOS element 8, a plurality of pixels are arranged along the spectral axis (x-axis direction) of AWG 1 corresponding to a plurality of WDM communication channels (for example, 10 channels). The TODC block 103 has a reflective configuration, and each optical signal emitted from the AWG 1 is reflected after being given a predetermined phase in the LCOS element 8 and returns to the AWG 1 again.

  The AWG 1 of the present TODC block has two input / output waveguides 16 and 17 that can input and output optical signal groups of a plurality of different wavelength groups. The first input / output waveguide 16 is connected to the slab waveguide 3 at a point on the boundary surface B opposite to the connection surface of the slab waveguide 3 to the arrayed waveguide 4. The first input / output waveguide 16 is connected to the first optical circulator 11 via an optical fiber or the like. The first optical circulator 11 functions as a port A, and an optical signal of a predetermined wavelength group is input (Ain) and output (Aout). Similarly, the second input / output waveguide 17 is connected to the slab waveguide 3 at a point b on the boundary surface B of the slab waveguide 3. The second input / output waveguide 17 is connected to the second optical circulator 14 via an optical fiber or the like. The second optical circulator 14 functions as a port B, and an optical signal having a wavelength group different from that of the port A is input (Bin) and output (Bout).

  Note that the points a and b to which the two input / output waveguides 16 and 17 are connected are shifted by a distance corresponding to 1/2 of the FSR set in the AWG 1. Therefore, the transmission band of the optical signal input / output from the port A and the transmission band of the optical signal input / output from the port B are shifted by 0.5 FSR on the optical frequency axis.

  FIG. 4 is a diagram illustrating the light transmittance of the TODC block according to the second embodiment. The light transmittance of the optical signal input / output using the port A is indicated by a solid transmittance line 21, and the light transmittance of the optical signal input / output using the port B is the dotted line transmittance. This is indicated by the line 22. Since all the transmittance lines have the same AWG transmittance, the transmittance characteristics with the FSR as a repetition period are shown. Note, however, that the center of the transmission band is shifted by 0.5 FSR between the two ports.

  In the transmittance characteristic by the port A, a flat portion at the center of the transmission band can be allocated for use of the optical signal of the wavelength group G1. In order to use the optical signal of the wavelength group G2 adjacent to the wavelength group G1, a flat portion at the center of the transmission band in the transmittance characteristic by the port B can be used. Further, for the wavelength group G3 adjacent to the wavelength group G2, in the transmittance characteristics by the port A, a flat portion at the center of the next transmission band that is FSR away from the transmission band used for the wavelength group G1 can be used. Similarly, for the wavelength group G4 adjacent to the wavelength group G3, the flat portion at the center of the next transmission band that is FSR away from the transmission band used for the wavelength group G2 can be used again in the transmittance characteristics by the port B.

  As described above, the ports used in the TODC block shown in FIG. 3 are alternately selected for each wavelength group, so that the light transmission characteristics are flat for a continuous communication band including a number of WDM communication channels. Dispersion compensation can be realized. At this time, the FSR set in the AWG 1 in the TODC block is twice the total bandwidth of the channels in one wavelength group. In the first embodiment, as shown in FIG. 2, the AWG FSR and the total bandwidth of the channels in one wavelength group are set to be the same. That is, in the first embodiment, the AWG FSR value of the TODC block is set to match the wavelength separation interval of the group demultiplexing filters 101 and 102. On the other hand, in Example 2, it should be noted that the AWG FSR value of the TODC block is set to twice the separation interval of the group demultiplexing filter. Next, the overall configuration of the tunable dispersion compensator using the TODC block shown in FIG. 3 will be further described.

FIG. 5 is a block diagram showing the overall configuration of the tunable dispersion compensator according to the second embodiment of the present invention. As in the first embodiment, an optical signal obtained by multiplexing optical signals having 40 wavelengths from λ ₁ to λ ₄₀ is divided into four wavelength groups (G1, G2, G3, G4). Dispersion compensation is performed using four TODC blocks. The second embodiment is different from the first embodiment in that the ports connected to the two group demultiplexing filters in the TODC block are alternately selected for each wavelength group between the A port and the B port. .

An optical signal in which an optical signal group having 40 wavelengths from λ ₁ to λ ₄₀ is multiplexed is input to the input port in of the first group demultiplexing filter 201. The multiplexed optical signal is converted into four wavelength groups (G1, G2) from λ ₁ to λ ₁₀ , λ ₁₁ to λ ₂₀ , λ ₂₁ to λ ₃₀ and λ ₃₁ to λ ₄₀ by the first group demultiplexing filter 201. , G3, G4). The optical signal group of each wavelength group output from the port 1, port 2, port 3, and port 4 of the wavelength group demultiplexing filter 201 is sent to the TODC blocks 203a, 203b, 203c, and 203d via optical fibers, respectively. Entered. The optical signals of each wavelength group are dispersion-compensated by the TODC blocks 203a, 203b, 203c, and 203d, respectively, and input to the respective ports of the second wavelength group demultiplexing filter 202. The optical signal groups of the four wavelength groups (G1, G2, G3, G4) that have been dispersion-compensated are multiplexed by the second wavelength group demultiplexing filter 202, and each of them has a central wavelength from λ ₁ to λ ₄₀ again. It is multiplexed into an optical signal including optical signal groups of 40 communication channels.

  The port A is used in the TODC 203a that performs dispersion compensation on the optical signal group of the wavelength group G1. The port B is used in the TODC 203b that performs dispersion compensation for the optical signal group of the wavelength group G2 adjacent to the wavelength group G1. Further, the port A is used again in the TODC 203c that performs dispersion compensation for the optical signal group of the wavelength group G3 adjacent to the wavelength group G2. In the TODC 203d that performs dispersion compensation for the optical signal group of the wavelength group G4 adjacent to the wavelength group G3, the port B is used again. As described above, the A port and the B port are used alternately for each wavelength group from G1 to G4. As a result, as shown in FIG. 4, it is understood that the transmission bands corresponding to the port A and the port B are alternately used for the adjacent wavelength groups. At this time, as the chromatic dispersion compensator, only the central region having a flat light transmittance is used in one transmission band corresponding to the FSR on the optical frequency axis. For this reason, only a flat central region is used without using a region corresponding to both ends of the FSR and being at both ends of the transmission band where the light transmittance decreases.

  According to the configuration of the second embodiment, the multiplexed WDM optical signal is separated into a plurality of wavelength groups, and the LCOS element having a short size in the direction of the spectral axis is used by including a plurality of one type of TODC blocks. be able to. A multi-channel variable dispersion compensator excellent in mass productivity and cost can be realized without requiring a large LCOS element. By setting the AWG FSR value of the TODC block to be twice the separation interval of the group demultiplexing filter, a variable dispersion compensator with a flat light transmission band can be realized. It is possible to reduce the cost of the tunable dispersion compensator, simplify the maintenance, and reduce the cost.

  In Example 2 described above, the connection point between the two input / output waveguides and the slab waveguide was shifted by a distance corresponding to 1/2 of the FSR set in the AWG 1. However, if the same technical idea is applied, a configuration including three input / output waveguides connected to the slab waveguide at a position shifted by a distance corresponding to 1/3 of the FSR can be provided. In this case, the TODC block includes three ports, port A, port B, and port C. It can be easily understood that a flat portion of light transmittance can be used in one wavelength group by dividing a plurality of consecutive wavelength groups into three groups that are alternately repeated. At this time, the AWG FSR value of the TODC block is set to three times the separation interval of the group demultiplexing filter. Furthermore, it is possible to expand the number of input / output waveguides and the number of ports to 3 or more based on the same concept.

  In the above-described first and second embodiments, a one-dimensional LCOS element in which pixels are arranged only in the AWG spectral axis direction is used as the LCOS element. However, by arranging the pixels two-dimensionally on the LCOS element and introducing the second spectroscopic element into the TODC block, the configuration of the tunable dispersion compensator can be further simplified and the LCOS element can be reduced in size. First, a TODC block including an LCOS element in which pixels are two-dimensionally configured will be described.

  FIG. 6 is a diagram illustrating the configuration of the TODC block according to the third embodiment of the present invention. Similar to the first and second embodiments, the configuration is a reflection type, but is different in that it further includes a bulk diffraction grating 15 and demultiplexes the optical signal also by the bulk diffraction grating. Furthermore, the present embodiment has a great feature in that when viewed through the optical path, the demultiplexing surfaces of the AWG and the bulk type diffraction grating are orthogonal to each other. Hereinafter, a detailed description will be given focusing on the difference from the TODC block of the first and second embodiments.

  Similar to the TODC block of the first and second embodiments, the optical signal input from the input fiber 10 enters the input waveguide 2 of the AWG 1 via the circulator 11 and the connecting fiber 13. The optical signal incident on the input waveguide 2 propagates to the arrayed waveguide 4 through the slab waveguide 3. In the arrayed waveguide 4, optical signal groups having different wavelengths are demultiplexed. That is, the optical signal emitted from the emission end 5 propagates toward the bulk diffraction grating 15 in the z-axis direction at different emission angles according to the wavelength in the xz plane (demultiplexing plane).

  The optical signal emitted from the emission end 5 is converted into a parallel beam by the cylindrical lens 6 in the thickness direction of the AWG substrate, that is, the y direction. The optical signal emitted from the cylindrical lens 6 can be regarded as parallel light in the yz plane. On the other hand, the x direction in the plane of the AWG substrate is converted into a sufficiently wide parallel beam by the lens action of the slab waveguide 3. That is, when the light is emitted from the emission end 5, the optical signal emitted from the AWG 1 can be regarded as parallel light in the xz plane. By passing through the cylindrical lens 6, the optical signal can be regarded as parallel light in both the x direction and the y direction.

  The configuration of the TODC block according to the third embodiment is characterized in that it includes a bulk diffraction grating that further demultiplexes the optical signal demultiplexed by the AWG 1. The optical signal emitted from the cylindrical lens 6 is further demultiplexed by the bulk diffraction grating 15 whose normal is inclined by θi with respect to the z-axis and the grating vector is set in the yz plane. The optical signal demultiplexed by the bulk diffraction grating 15 is condensed on the spatial phase control element 8 by the condenser lens 7. Here, the dispersion direction of the AWG 1 and the dispersion direction of the bulk diffraction grating 15 are in a relationship in which the two demultiplexing surfaces are orthogonal to each other when the demultiplexing surfaces are viewed along the optical path.

  According to FIG. 6A, the AWG 1 and the spatial phase control element 8 are described so as to be arranged in parallel with each other, but strictly speaking, they need not be parallel. FIG. 6 exemplarily shows a case where a specific bulk diffraction grating described later is used and the incident angle θi is 46.76 °. At this time, the optical path is refracted at about 90 ° in the bulk diffraction grating. For this reason, in the drawing, the AWG and the spatial phase control element are expressed as if they were arranged in a parallel positional relationship. Therefore, in this embodiment, there is no limitation on the refraction angle θi of the bulk type diffraction grating. The TODC block of the present embodiment has a spatial phase control element pixel as a result of the relationship between the demultiplexing surface of AWG1 and the demultiplexing surface of bulk diffraction grating 15 being relatively orthogonal when viewed along the optical path. It is characterized in that it can be configured in two dimensions with anisotropy.

  To explain the relationship between the wavelength (optical frequency) and the position of the focused beam of the optical signal, consider the locus that the focused beam draws on the spatial phase control element when the wavelength is virtually continuously changed. Try. In this embodiment, the angular dispersion of the AWG 1 is set to be sufficiently larger than the angular dispersion of the bulk diffraction grating 15, so that the condensed beam on the spatial phase control element 8 is scanned in a raster shape according to the wavelength of the optical signal. The

  For example, the diffraction order of the bulk diffraction grating 15 may be set to 1, and the FSR of the AWG 1 may be set to be equal to the entire bandwidth of the WDM communication channel of one wavelength group in the communication system to be subjected to dispersion compensation. Such a raster scan of the beam can be performed by using an AWG that has a high degree of freedom in design parameters and can easily realize a desired FSR as the first spectroscopic element. Even if a bulk diffraction grating is used as the first spectroscopic element, a desired angular dispersion cannot be set easily. Note that dispersion compensation operation peculiar to this embodiment is realized by allocating and combining appropriate angular dispersions to the demultiplexing characteristics of the first and second spectroscopic elements. I want to be. Similar to the TODC blocks of the first and second embodiments, the optical signal reflected by the spatial phase control element 8 reverses its optical path and propagates in the direction opposite to the forward path, and passes through the circulator 11 to the output fiber 12. It is emitted from.

  FIG. 7A is a diagram illustrating a configuration of a spatial phase control (LCOS) element suitable for the TODC block according to the third embodiment. The coordinate system on the spatial phase control element 8 is defined as u axis-v axis. As described above, the FSR of AWG 1 is set equal to the entire bandwidth of the WDM communication channel in one wavelength group in the communication system that is the object of dispersion compensation. At this time, an optical signal in the entire band of one wavelength group corresponds to interference light of a certain diffraction order of AWG1. Assuming that the wavelength of the virtually unmodulated optical signal is continuously changed, the position of the focused beam draws a locus on the line segment Lm for the optical signal of diffraction order m by the demultiplexing action of AWG1. .

  This optical signal of diffraction order m corresponds to, for example, an optical signal group in the mth specific wavelength group. Similarly, for an optical signal of diffraction order m + 1, a locus on adjacent Lm + 1 is drawn. This optical signal of diffraction order m + 1 corresponds to an optical signal group in the (m + 1) th wavelength group adjacent to the mth wavelength group. Therefore, the optical signal component within the entire band of one wavelength group corresponds to a pixel column localized on one locus line Lm that is scanned and drawn on the spatial phase control element 8. In other words, each optical signal within the entire band of one wavelength group is independently phased by a plurality of pixel rows arranged in the u-axis direction localized on one locus Lm, and dispersion compensation is performed. Is realized.

  Dispersion characteristics (group delay characteristics) can be set for different wavelength groups for each pixel array arranged in the u-axis direction. That is, the TODC block according to the third embodiment has a characteristic that different dispersion characteristics (group delay characteristics) can be set for each wavelength group by setting different phase distributions independently for each pixel column. In FIG. 6, the direction of the line of intersection between the spectral plane (ie, yz plane) of the bulk type diffraction grating 15 and the pixel formation surface of the spatial phase control element 8 is the dispersion direction by the bulk type diffraction grating 15 on the spatial phase control element 8. It becomes. The dispersion axis defined by this intersection line is taken as the z ′ axis. In FIGS. 7A and 7B, the z′-axis is a direction connecting the end points of the locus lines such as Lm and Lm + 1 or a direction connecting the start points.

With reference to FIG. 7A, the relationship between the focused beam diameter and the pixel structure formed on the u-axis and v-axis planes will be further examined. In FIG. 7A, the shape of each pixel is displayed as a square for simplicity. In the following, it should be noted that the method of setting the phase for each pixel will be described by focusing on the relative relationship between the focused spot beam radius and the pixel pitch in each of the u axis and the v axis. In addition, the shape of the condensed beam corresponding to an optical signal having a certain optical frequency (wavelength) that is not modulated is generally an ellipse according to the characteristics of the condensing lens and the cylindrical lens. Here, the ellipse radius in the v-axis direction is wv. The ellipse radius refers to a radius at which the light intensity of the focused spot is 1 / e ^{2 of the} peak value, that is, a radius at which the light intensity is 13.5% of the peak light intensity.

In the v-axis direction, the condensing spot raster that describes the locus Lm moves dv for each optical frequency corresponding to the FSR of the arrayed waveguide grating based on the angular dispersion of the bulk diffraction grating 15. Therefore, overlapping of beams of adjacent wavelength groups can be eliminated by satisfying the following equation for the elliptical radius of the focused spot in the v-axis direction.
wv ≦ dv / 2 Formula (1)
Since the elliptical radius wv in the v-axis direction is a radius at which the light intensity of the focused spot is 1 / e ^{2 of the} peak value, satisfying the condition of the formula (1), the −30 dB or less generally required in optical communication. Crosstalk performance can be realized.

  In the configuration of FIG. 7A, the direction of the locus Lm determined by the linear dispersion values of the AWG 1 and the bulk diffraction grating 15 is made to coincide with the u-axis direction of the spatial phase modulation element 8. Furthermore, the pixel pitch in the v-axis direction is set to pSLMv, and pSLMv and dv are matched. With the above-described pixel configuration, each optical signal in the entire band of one wavelength group can correspond to one pixel column aligned in the u-axis direction. As a result, dispersion can be imparted to the optical signals in all wavelength groups using the LCOS having the minimum number of pixels. By reducing the number of pixels, the cost of LCOS can be kept low.

  Below, the example of the TODC block of a present Example is shown with a specific numerical example. The arrayed waveguide grating was fabricated using a silica-based optical waveguide having a relative refractive index difference of 1.5%. The path length difference ΔL of the arrayed waveguide was 202 μm, and the arrayed waveguide pitch at the exit end 5 of the arrayed waveguide was 12 μm. According to this configuration, the free spectral range of the arrayed waveguide grating is approximately 1000 GHz.

  The bulk type diffraction grating 15 uses, for example, a volume phase holographic grating (VPHG) having a grating period of 940 lines / mm. The bulk type diffraction grating 15 is not limited to the VPHG type, and a function similar to that of the VPHG can be realized by using a transmission type blaze diffraction grating, a reflection type holographic diffraction grating, or a reflection type blaze diffraction grating. When the incident angle θi is 46.76 °, the angular dispersion value of VPHG with a grating period of 940 lines / mm is 1.37 mrad / nm. The focal length of the cylindrical lens 6 was 1 mm, and the focal length of the condenser lens 7 was 80 mm.

  In the LCOS type spatial phase control element 8, the number of pixels and the pitch in the u-axis direction are 1280 and 8 μm, respectively, and the number of pixels and the pitch in the v-axis direction are 4 and 920 μm, respectively. Therefore, the size of the region where the LCOS pixels are formed is about 10.2 mm × 3.7 mm. This configuration is merely an example, and the pixel pitch in the u-axis direction may be in the range of 5 μm to 10 μm.

  According to the configuration of the optical system described above, the beam radius wv in the v-axis direction is about 300 μm, and it has been confirmed that the relationship of Expression (1) is satisfied. The linear dispersion value in the v-axis direction on LCOS is found to be 0.11 mm / nm as the product of the above-mentioned angular dispersion value of VPHG and the focal length of the condenser lens 7. Therefore, the position of the focused spot moves dv = 920 μm per 1000 GHz (about 8.4 nm) which is the FSR of the AWG. It was confirmed that both the spot movement amount dv corresponding to the FSR of the arrayed waveguide grating and the pixel pitch pSLMv in the v-axis direction were 920 μm.

  The linear dispersion value in the u-axis direction of the LCOS element is 1.22 mm / nm, and 128 pixels contribute to the phase modulation of an optical signal having a frequency range of 100 GHz. Therefore, the TODC block having the configuration of the present embodiment makes it possible to perform dispersion compensation independently for each optical signal of 10 WDM communication channels arranged at 100 GHz intervals in one wavelength group. Furthermore, dispersion compensation can be performed independently for a plurality of optical signal groups of different wavelength groups for each pixel array arranged in the v-axis direction.

  In the configuration example of the LCOS shown in FIG. 7A, the case where the pSLMv and dv are substantially matched and one pixel row arranged in the u-axis direction corresponds to one wavelength group is exemplified. explained. As another configuration example, a configuration example will be described in which dv is set larger than pSLMv and a plurality of pixels are assigned to one wavelength group in the v-axis direction, as shown in FIG. Also in this case, the crosstalk between adjacent wavelength groups can be kept low by satisfying the relationship of the expression (1). In the pixel configuration of FIG. 7A, it is necessary to make the direction of the locus Lm coincide with the direction of the u axis. On the other hand, in the pixel configuration of FIG. 7B, by using a plurality of pixels included in the width of dv centering on the locus Lm in the v-axis direction for controlling one wavelength group m, The locus Lm and the u axis need not be parallel.

  The advantage of the pixel configuration of FIG. 7B is that an arbitrary optical coupling loss can be added to the optical signal. A plurality of pixels in the v-axis direction can set a phase tilted with respect to the v-axis direction independently of the phase setting with respect to the u-axis direction. Referring to FIG. 6 again, the optical signal reflected by the LCOS and passed through the bulk diffraction grating 15 is tilted with respect to the AWG waveguide eigenmode in the y-axis direction at the output end 5 of the arrayed waveguide grating. The light enters the AWG 1 with a distribution. Therefore, it is possible to add a wavelength-dependent loss to the transmission characteristics of the TODC block and compensate the wavelength dependence of the optical signal intensity.

  As an LCOS for realizing the configuration of FIG. 7B, an LCOS in which pixels are arranged on a general square lattice was used. The pixel pitch is 8 μm in both the u-axis direction and the v-axis direction. The configuration of the optical system other than LCOS was the same as that described with reference to FIG. Since dv = 920 μm, 115 pixels are assigned to one wavelength group in the v-axis direction around the locus Lm. At any wavelength, that is, at any wavelength group number m and any position on the u-axis, the phase value given to 115 pixels assigned in the v-axis direction is a phase change with a maximum inclination angle of 0.3 degrees. Was linearly changed in the v-axis direction corresponding to. With this linearly tilted phase, the optical signal intensity at that wavelength could be controlled in the range of 0 dB to -40 dB.

  The variable dispersion compensator according to the first embodiment shown in FIG. 1 is obtained by using a TODC block including LCOS elements in which pixels shown in FIG. 7A or 7B are two-dimensionally arranged. It can be modified to a simpler configuration.

  In the above description related to LCOS, the pixel width and the space between the pixels are not touched by paying attention only to the pixel pitch, but the space between the pixels is narrow to increase the light control efficiency regardless of the usage. It is preferable to do. In general, 1 μm or less is preferable.

  FIG. 8 is a block diagram illustrating the overall configuration of the dispersion compensator according to the third embodiment. FIG. 8 represents the dispersion compensator divided into blocks focusing on the function as the dispersion compensator. Hereinafter, the comparison relationship between the TODC block shown in FIG. 6 and a specific configuration will be described. The LCOS element 8 in the TODC block of FIG. 6 has a configuration in which pixels are arranged in two dimensions (2-Dimension) as shown in FIGS. 7A and 7B. Called an element.

In the tunable dispersion compensator 300 of FIG. 8, for example, an optical signal in which an optical signal group having a wavelength of 40 from λ ₁ to λ ₄₀ is multiplexed is input to the input port In of the spectroscope 301 and demultiplexed. And input to the first group demultiplexing filter 302. The multiplexed optical signal is divided into four wavelength groups (G1, G2,..., Λ ₁ to λ ₁₀ , λ ₁₁ to λ ₂₀ , λ ₂₁ to λ ₃₀ and λ ₃₁ to λ ₄₀ by the first group demultiplexing filter 302. G3, G4) are group-demultiplexed. The optical signal group of each wavelength group from the first wavelength group demultiplexing filter 302 is a pixel group 306a in each column (first column, second column, third column, fourth column) on the 2D-LCOS element 305. , 306b, 306c, and 306d. The optical signal groups of the four wavelength groups (G1, G2, G3, G4) after dispersion compensation are multiplexed by the second wavelength group demultiplexing filter 303 and the spectroscope 304, and again centered from λ ₁ to λ _40. The signals are multiplexed into the optical signals of 40 communication channels having wavelengths and output from the output port out.

  The two spectrometers 301 and 304 in FIG. 8 correspond to AWG1 in the TODC block in FIG. Here, since the TODC block shown in FIG. 6 has a reflection type configuration, it should be noted that the functions of the spectrometer 301 and the spectrometer 304 can be realized by one AWG. The two group demultiplexing filters 302 and 303 in FIG. 8 correspond to the bulk diffraction grating 15 (second spectroscopic means) in the TODC block shown in FIG. Similar to the AWG 1, the functions of the two group demultiplexing filters 302 and 303 can be realized by one bulk diffraction grating 15.

  The 2D-LCOS element 305 includes a first column of pixel groups 306a corresponding to the wavelength group G1, a second column of pixel groups 306b corresponding to the wavelength group G2, a third column of pixel groups 306c corresponding to the wavelength group G3, and a wavelength. Dispersion compensation is performed by the pixel group 306d in the fourth column corresponding to the group G4. Here, in the 2D-LCOS element, pixel groups in each column (first column, second column, third column, and fourth column) arranged in the u-axis direction within one element plane are two-dimensionally integrated. Note that the pixels are arranged and arranged. Therefore, the optical signal processing from when the optical signal is emitted from the spectroscope 301 (AWG1) to the space, subjected to the optical signal processing by the bulk diffraction grating and the 2D-LCOS element, and again incident on the spectroscope 304 (AWG1) is performed. This is realized by a set of TODC blocks. That is, the same function as the tunable dispersion compensator that requires the four TODC blocks shown in FIG. 1 can be realized by one TODC block shown in FIG.

  Note that although it has been described that each pixel column of the 2D-LCOS element 305 is from the first column to the fourth column, phase setting is performed using a plurality of pixels in the v-axis direction for one wavelength group. Note that you can also. In this case, a plurality of pixel columns arranged in the u-axis direction are arranged for one wavelength group.

  According to the tunable dispersion compensator according to the first embodiment shown in FIG. 1, separate wavelength group demultiplexing filters 101 and 102 are required in addition to the four TODC blocks. On the other hand, since the tunable dispersion compensator 300 according to the third embodiment can be configured with only one TODC block shown in FIG. 6, the configuration of the tunable dispersion compensator can be greatly simplified.

  Similar to the tunable dispersion compensator of the first embodiment, the tunable dispersion compensator of the second embodiment shown in FIG. 5 can be modified to a simpler configuration by using the 2D-LCOS element shown in FIG. Can do. That is, in the TODC block using the LCOS element in which the pixels are two-dimensionally shown in FIG. 6, the two input / output waveguides 16 and 17 connected to the AWG as in FIG. It is possible to realize variable dispersion compensation having a flat light transmission characteristic for a continuous communication band (wavelength group) including a WDM communication channel.

  FIG. 9 is a configuration diagram of a TODC block used in the variable dispersion compensator of the fourth embodiment. The configuration shown in FIG. 9 is that the second demultiplexing means (bulk type diffraction grating) 15 is added to the TODC block (FIG. 3) used in the variable dispersion compensator of the second embodiment. This is different from the configuration of the TODC block shown in FIG. Further, in terms of having two input / output waveguides 16 and 17 connected to a position equivalent to 0.5 FSR on the boundary surface of the slab waveguide 3, it is common to the configuration of the TODC block of FIG. The difference is that the interleave type group demultiplexing filter 19 is provided. Hereinafter, the description will be given focusing on these differences. In FIG. 9, for simplicity, the optical path bending by the second demultiplexing means 15 is omitted and is shown in a simplified manner.

The TODC block shown in FIG. 9 has a reflective configuration. An optical signal in which an optical signal group having 40 wavelengths from λ ₁ to λ ₄₀ is multiplexed is input to the TODC block via the optical circulator 11. After being compensated for dispersion in the TODC block, it is output from the optical circulator 11 again. The multiplexed input optical signal to the optical circulator 11 is input to the interleave type group demultiplexing filter 19. The interleave type group demultiplexing filter 19 demultiplexes the multiplexed optical signal group including a plurality of wavelength groups into an odd-numbered wavelength group and an even-numbered wavelength group. That is, two wavelength groups (G 1 and G 3) from λ ₁ to λ ₁₀ and λ ₂₁ to λ ₃₀ are group-demultiplexed at the output on the port A side of the interleave type group demultiplexing filter 19. Two wavelengths (G2, G4) from λ ₁₁ to λ ₂₀ and from λ ₃₁ to λ ₄₀ are group-divided at the output on the port B side of the interleave type group demultiplexing filter 19. The optical signal groups from the A port and the B port are input to the two input / output waveguides 16 and 17 connected to the positions separated by 0.5 FSR on the boundary surface of the slab waveguide 3, respectively. Similar to the TODC of the second embodiment, the FSR value of the AWG 1 is set to twice the wavelength group separation interval of the interleave type group demultiplexing filter 19.

  The LCOS element 8 in the TODC block of FIG. 9 is a 2D-LCOS element in which pixels are arranged in a two-dimensional manner as shown in FIG. 7 as in the third embodiment.

  FIG. 10 is a block diagram illustrating an overall configuration of a dispersion compensator according to the fourth embodiment. FIG. 10 represents the dispersion compensator divided into blocks focusing on the function as the dispersion compensator. Hereinafter, description will be made in comparison with a specific configuration of the TODC block shown in FIG.

In the tunable dispersion compensator 400 of FIG. 10, for example, an optical signal obtained by multiplexing optical signal groups having 40 wavelengths from λ ₁ to λ ₄₀ is converted into an odd-numbered wavelength by an interleaved group demultiplexing filter 401. The interleaved group is demultiplexed into a group and an even-numbered wavelength group. The A wavelength group (G1, G3) and the B wavelength group (G2, G4), which are demultiplexed by the interleave group, are input to the A port and the B port of the spectrometer 402, respectively. Here, the spectroscope 402 corresponds to AWG1 in FIG. Further, as in the third embodiment, the first wavelength group demultiplexing filter 403 corresponds to the second demultiplexing means 15 in FIG. The optical signal group of each wavelength group from the first wavelength group demultiplexing filter 403 is a pixel group in each column (first column, second column, third column, fourth column) on the 2D-LCOS element 407. Dispersion compensation is performed by 408a, 408b, 408c, and 408d. The optical signal groups of the four wavelength groups (G 1, G 2, G 3, G 4) after dispersion compensation are combined by the second wavelength group demultiplexing filter 404 and the spectroscope 405. Further, the interleaved group demultiplexing filter 406 interleaves the odd-numbered wavelength group and the even-numbered wavelength group, and again multiplexes them into optical signals of 40 communication channels having center wavelengths from λ ₁ to λ _40. And output from the output port Out.

  Similarly to the third embodiment, the spectroscopes 402 and 405 and the group demultiplexing filters 403 and 404 are realized by one AWG 1 and one second demultiplexing unit 15 by the TODC block having the reflection type configuration of FIG. Needless to say, you can.

  The 2D-LCOS element 407 includes a first column of pixel groups 408a corresponding to the wavelength group G1, a second column of pixel groups 408b corresponding to the wavelength group G2, a third column of pixel groups 408c corresponding to the wavelength group G3, and a wavelength. Dispersion compensation is performed by the pixel group 408d in the fourth column corresponding to the group G4. Here, it should be noted that in the 2D-LCOS element 407, pixels are arranged in a two-dimensional manner in which pixel groups of each column are integrated in one element plane. Therefore, an optical signal is processed until an optical signal is emitted from the spectroscope 402 (AWG1) to the space, subjected to optical signal processing by the bulk diffraction grating 15 and the 2D-LCOS element, and again incident on the spectroscope 405 (AWG1). Is realized by a set of TODC blocks. That is, it should be noted that one TODC block shown by FIG. 9 can realize the same function as the tunable dispersion compensator shown in FIG.

  In the tunable dispersion compensator according to the second embodiment illustrated in FIG. 5, separate wavelength group demultiplexing filters 201 and 202 are required in addition to the four TODC blocks. On the other hand, since the variable dispersion compensator 400 of the fourth embodiment can be configured with only one TODC block shown in FIG. 9, the configuration of the variable dispersion compensator can be greatly simplified.

  Similar to the second embodiment, by separating the multiplexed WDM optical signal into a plurality of wavelength groups and including only one TODC block, an LCOS element having a short size in the spectral axis direction can be used. A multi-channel variable dispersion compensator excellent in mass productivity and cost can be realized without requiring a large LCOS element. By setting the AWG FSR value of the TODC block to be twice the wavelength group separation interval of the interleave type group demultiplexing filter, a variable dispersion compensator with a flat light transmission band can be realized. It is possible to reduce the cost of the tunable dispersion compensator, simplify the maintenance, and reduce the cost.

  As described above in detail, in the third and fourth embodiments, the number of TODC blocks is further reduced as compared with the variable dispersion compensator having the configuration of the first or second embodiment, so that a multi-channel variable dispersion compensator is obtained. Can be configured. By setting the FSR so as to have a predetermined relationship with the entire bandwidth of the WDM communication channel group covered by the TODC block, the light transmission characteristics can be flattened.

  As described in detail in Example 3 and Example 4, by using a 2D-LCOS element to form a two-dimensional pixel, an LCOS element in which the size in the direction corresponding to the spectral axis of the AWG is shortened can be obtained. The same effect as using it is obtained.

In the dispersion compensator of the second embodiment, the group demultiplexing filters 201 and 202 include four wavelength groups (G1, G2) of λ ₁ to λ ₁₀ , λ ₁₁ to λ ₂₀ , λ ₂₁ to λ ₃₀ and λ ₃₁ to λ _40. , G3, G4) were multiplexed / demultiplexed. In order to realize such an optical group demultiplexing, the group demultiplexing filter transmits the optical signals of λ ₁ to λ _{10 in} the band of the wavelength group G1 with low loss, and transmits the band of the adjacent wavelength group G2. A steep filter characteristic that cuts off optical signals from λ ₁₁ to λ ₂₀ is required. It is difficult to realize a group demultiplexing filter having such characteristics.

  Further, in a communication system using an interleaver, communication channels are used in a skipped manner, and dispersion compensation is performed to multiplex / demultiplex optical signal groups having adjacent wavelength groups arranged continuously as in the second embodiment. The vessel cannot be used as it is. Therefore, in this embodiment, a variable dispersion compensator using an interleaver as a group demultiplexing filter and having a flat light transmission band will be described. Note that in the following description, there is a one-to-one correspondence between channels and wavelengths. That is, selecting an optical signal having a specific wavelength from the viewpoint of the physical properties of the optical signal is the same as selecting a corresponding channel from the viewpoint of the function for the corresponding communication channel.

  FIG. 11 is a block diagram illustrating an overall configuration of a tunable dispersion compensator according to the fifth embodiment of the present invention. The tunable dispersion compensator 600 of the present embodiment includes the same components as those of the tunable dispersion compensators of the first embodiment shown in FIG. 1 and the second embodiment shown in FIG. 5, and has the same overall configuration. However, the present embodiment is different from the first and second embodiments in that the TODC block is configured as a reflection type configuration and as a reflection type variable dispersion compensator. The tunable dispersion compensator 600 includes a group demultiplexing filter 601 that demultiplexes multiplexed optical signals and TODC blocks 603a, 603b, 603c, and 603d that perform dispersion compensation on the group demultiplexed optical signals. The group demultiplexing filter 601 is connected to an optical fiber 602 to be subjected to dispersion compensation. If the TODC block has a transmission type configuration, a transmission type variable dispersion compensator can also be configured. In this case, two group demultiplexing filters are included. The following description will be given focusing on differences from the configuration of the dispersion compensator of the second embodiment.

In the present embodiment, an interleaved group demultiplexing filter is used as the group demultiplexing filter 601 for group demultiplexing of multiplexed optical signals. Hereinafter, the interleave type group demultiplexing filter in this embodiment is called an interleaver for simplicity. In order to configure a dispersion compensator for 44 channels (channel numbers ₁ to ₄₄₎ , the interleaver 601 in this embodiment uses an optical signal in which optical signals having 44 wavelengths from λ ₁ to λ ₄₄ are multiplexed. Group splitting into four wavelength groups (G1, G2, G3, G4). In the present embodiment, the wavelength of each of the demultiplexed wavelength groups is G1 (λ ₁ , λ ₅ , λ ₉ ,... Λ ₃₇ , λ ₄₁ ), G2 (λ ₂ , λ ₆ , λ ₁₀ ,. Λ ₃₈ , λ ₄₂ ), G ₃ (λ ₃ , λ ₇ , λ ₁₁ ,... Λ ₃₉ , λ ₄₃ ), G 4 (λ ₄ , λ ₈ , λ ₁₂ ,... Λ ₄₀ , λ ₄₄ ) This is different from the operation of the group demultiplexing filter in the first and second embodiments. That is, the interleaver 601 demultiplexes or multiplexes all target wavelength groups (here, wavelength groups corresponding to 44 channels) into a predetermined number of wavelength groups b (4). It is assumed that each wavelength (channel) in one wavelength group is selected at equal channel intervals. For example, the wavelength group G1 of this embodiment has channel numbers 1, 5, 9,... 37, 41, and is therefore selected at intervals of 4 channels.

  FIG. 12 is a diagram illustrating a configuration including a more detailed configuration of the TODC block of the tunable dispersion compensator according to the fifth embodiment. 11 shows a more detailed configuration of each TODC block 603a, 603b, 603c, and 603d in FIG. A circulator 607 having an input port in and an output port out is provided on the optical fiber side of the group demultiplexing filter 601, and an optical signal is input and output in a dispersion compensator having a reflection type configuration. In addition, the structure which equips each TODC block with a circulator is also possible.

Hereinafter, the TODC block 603a will be described as an example. The TODC block 603a includes a second group branching filter 604a and a TODC 605a. The second group demultiplexing filter 604a further multiplexes and demultiplexes the wavelength group G1 group-demultiplexed by the interleaver 601 into two sub-wavelength groups g (1-1) and g (1-2). The sub-wavelength group g (1-1) includes λ ₁ , λ ₅ , λ ₉ , λ ₂₅ , λ ₂₉ , and λ ₃₃ , and the sub-wavelength group g (1-2) includes λ ₁₃ , λ ₁₇ , and λ _21. , Λ ₃₇ , and λ ₄₁ . The filter characteristics of the second group branching filter will be described later.

  The optical signals of the two sub-wavelength groups g (1-1) and g (1-2) of the second group demultiplexing filter are connected to the A port and the B port of the TODC 605a, respectively. The TODC block 605a has the same configuration as the TODC block of the second embodiment shown in FIG. 5, and includes an AWG, a condenser lens, a spatial phase modulation element (LCOS), and the like. As will be described later with reference to FIG. 17, the TODC block 603a includes two input / output waveguides connected to the slab waveguide at each position corresponding to a difference of 0.5 FSR. These two input / output waveguides correspond to two input / output ports (A, B) of the TODC block, as will be described later.

  The other TODC blocks 603b, 603c, and 603d have the same configuration and perform the same operation. That is, the TODC block 603b includes a second group demultiplexing filter 604b and a TODC 605b. The TODC block 603c includes a second group branching filter 604c and a TODC 605c. Further, the TODC block 603d includes a second group branching filter 604d and a TODC 605d. The second group demultiplexing filter 604b further multiplexes and demultiplexes the wavelength group G2 group-demultiplexed by the interleaver 601 into two sub-wavelength groups g (2-1) and g (2-2). The second group demultiplexing filter 604c multiplexes and demultiplexes the wavelength group G3 group-demultiplexed by the interleaver 601 into two sub-wavelength groups g (3-1) and g (3-2). Then, the second group demultiplexing filter 604d multiplexes and demultiplexes the wavelength group G4 group-demultiplexed by the interleaver 601 into two sub-wavelength groups g (4-1) and g (4-2). .

  Since the dispersion compensator of the fifth embodiment has a reflection type configuration, the TODC blocks 603a, 603b, 603c, and 603d have a reflection type configuration. Accordingly, only one second group demultiplexing filter 604a is provided in the TODC block 603a. However, in the case of a transmission type configuration, two second group demultiplexing filters can be included.

  As described above, the interleaver (first group demultiplexing filter) demultiplexes all the target wavelength groups (here, the wavelength groups corresponding to 44 channels) into a predetermined number of wavelength groups (4). Or combine. Wavelengths (channels) within one wavelength group are selected at equal intervals.

  The second group demultiplexing filter has a characteristic of separating into continuous wavelength groups having the maximum number of channels that can be processed by one TODC block of the tunable dispersion compensator of the present invention. More specifically, with respect to the optical signal group of the channel with the total number of channels N = 44, the optical signal group is transmitted or blocked every 11 adjacently arranged channels. This characteristic is the same as the group demultiplexing filter characteristic of the group demultiplexing filter described in the first and second embodiments. If this characteristic is used, two sub-wavelength groups are alternately arranged for every 11 channels. It is necessary to have a group branching filter characteristic that branches into Therefore, as described at the beginning of this embodiment, there is a problem that it is difficult to realize steep filter characteristics in the group demultiplexing filter. However, as in this embodiment, this problem can be avoided by combining an interleaver and a group demultiplexing filter having a filter characteristic that multiplexes and demultiplexes each adjacent channel arranged continuously. The second group demultiplexing filters 604a, 604b, 604c, and 604d in each TODC block 603a, 603b, 603c, and 603d can be realized with a gentle transmission band as will be described below. Each second group demultiplexing filter has a filter characteristic corresponding to the wavelength group of each TODC block, and the center wavelength of the transmission band is substantially shifted by one channel.

  In general, the second group demultiplexing filter can be realized by a thin film filter (hereinafter referred to as TFF) that can be produced by stacking a plurality of deposited films. It is known that more layers need to be formed in order to produce a filter having a steep wavelength characteristic (see Non-Patent Document 4). By relaxing the required characteristics of the edge portions of the filter in the transmission region and the cutoff region, the manufacturing process of the TFF is simplified and the manufacture becomes easy.

  In the variable dispersion compensator of this embodiment, an example using LCOS as the spatial phase modulation element will be described. In this embodiment, as described below, the number of LCOS element elements (cells) corresponding to one channel (wavelength) is associated with the AWG FSR, and further selected discretely and periodically. By assigning a channel to each cell of LCOS, it is possible to avoid the influence of a decrease in light transmittance at both ends of one transmission band corresponding to FSR on the optical frequency axis. In the following description, a cell refers to a group of element elements corresponding to signal processing of one communication channel. Taking LCOS as an example, among a plurality of pixels arranged in the direction of the demultiplexing axis, a group of liquid crystal pixels in one section corresponds to a cell. In the following description, it should be noted that signal processing for one communication channel is performed by one cell.

  FIG. 13 is an overview diagram showing the channel number and LCOS cell assignment and the corresponding TFF filter characteristics in the variable dispersion compensator of this embodiment in association with each other. FIG. 13A shows channel (CH) numbers assigned to the TODC blocks 603a, 603b, 603c, and 603d. Hereinafter, for simplicity, each TODC block is referred to as TODC1, TODC2, TODC3, and TODC4. As shown in FIG. 12, the wavelength group G1 is assigned to TODC1, the wavelength group G2 is assigned to TODC2, the wavelength group G3 is assigned to TODC3, and the wavelength group G4 is assigned to TODC4.

  FIG. 13B shows the filter characteristics of the second group demultiplexing filter corresponding to each TODC block. Hereinafter, the second group demultiplexing filter is referred to as TFF for simplicity. At the top, the ideal TFF filter characteristics are shown. Below that, the TFF characteristics for TODC1 to TODC4 of this embodiment are shown as TFF1, TFF2, TFF3, and TFFF4, respectively. The ideal characteristics of TFF are required to transmit optical signals of channel numbers 1 to 11 and to block optical signals of channel numbers 12 to 22. It has been difficult to realize such a steep filter characteristic. However, the TFF in the present embodiment may have a gradual filter characteristic corresponding to each wavelength group, such as the filter characteristics shown as TFF1 to TFF4. In each TFF characteristic, a characteristic indicated by a solid line is a filter characteristic when connected to one (A port) of two input / output ports of the AWG described later with reference to FIG. The characteristic indicated by a broken line is a corresponding filter characteristic when connected to the other (B port) of the input / output ports of the two AWGs. The filter characteristics of TFF1 to TFF4 are shifted by about one channel in the wavelength axis (frequency axis) direction.

  FIG. 13C shows the LCOS cell numbers to which channels are assigned for each of TODC1 to TODC4 in association with the channel number of (A) and the filter characteristics of TFF of (B). As the spatial phase modulation element in the TODC block of this embodiment, LCOS is typically used. The number of cells included in each LCOS is eleven. In (C), the upper number indicates the cell number when connected to the A port, and for TODC1, the channel number of the sub-wavelength group g (1-1) and the cell number are associated with each other. The lower number indicates the cell number when connected to the B port. For TODC1, the channel number of the sub-wavelength group g (1-2) and the cell number are associated with each other. With the schematic view of FIG. 13, how channels are allocated to LCOS cells in this embodiment will be described in further detail.

FIG. 14 is a diagram for explaining channel assignment for LCOS cells in the tunable dispersion compensator of the fifth embodiment. (A) shows the relationship between the LCOS cell number and channel number for TODC1. (B) to (D) show the relationship between cell numbers and channel numbers for TODC2 to TODC4, respectively. Referring to (A), LCOS cell numbers 1 to 11 are shown in the upper part. The lower part shows an allocation table in which channel numbers 1 to 44 are sequentially arranged while being folded back every 11 channels. Channel numbers of 11 wavelengths of the wavelength group G1 (λ ₁ , λ ₅ , λ ₉ , λ ₁₃ , λ ₁₇ , λ ₂₁ , λ ₂₅ , λ ₂₉ , λ ₃₃ , λ ₃₇ , λ ₄₁ ) assigned to the TODC ₁ The columns are hatched. Since the channel numbers are selected sequentially at intervals of 4 channels for 11 cells, the channel numbers are shifted one by one for each column of the allocation table of (A), that is, the channels for different cells. Is selected.

In addition, the channel numbers of the wavelengths λ ₁ , λ ₅ , λ ₉ , λ ₂₅ , λ ₂₉ , and λ ₃₃ of the sub-wavelength group g (1-1) are selected for the channels in the first and third columns of the table. The channel numbers of the wavelengths λ ₁₃ , λ ₁₇ , λ ₂₁ , λ ₃₇ , and λ ₄₁ of the sub-wavelength group g (1-2) are selected for the channels in the second and fourth columns of the table. . Therefore, the channels selected in the first and third columns correspond to the channels selected when connected to the A port in the AWG. The channels selected in the second and fourth rows correspond to the channels selected when connected to the B port in the AWG.

  FIG. 15 is another diagram illustrating channel assignment for LCOS cells in the tunable dispersion compensator of the fifth embodiment. It is the figure which simplified and demonstrated the channel allocation with respect to the LCOS cell of each TODC based on the table | surface shown in FIG. (A) describes the channel number assigned to each cell of the LSOC when connected to each of the A port and the B port in the TODC 1. That is, the channels selected in the first and third columns in the table of FIG. 14A are described in one column as the selected channel corresponding to the A port. Similarly, the channels selected in the second and fourth columns in the table of FIG. 14A are described in one column as the selected channel corresponding to the B port. 14B to 14D also show the channel numbers assigned to each cell of the LSOC when the TODC2 to TODC4 are connected to the A port and the B port, respectively. As is apparent from FIG. 15, in any TODC, the channel number when the A port is connected and the channel number when the B port is connected are alternately assigned to all 11 cells of the LCOS without any gap. .

  FIG. 16 is a diagram illustrating channel assignment results for LCOS cells in the tunable dispersion compensator of the fifth embodiment. Based on FIG. 15, the channel numbers assigned to the LCOS cells of each TODC are shown. For example, referring to (A), all 11 wavelengths (channels) of the wavelength group G1 for TODC1 do not overlap or become unused for all 11 cells of LCOS, and there is no excess or deficiency. Assigned.

In the present embodiment, the efficient and wasteful channel assignment as described above can be performed first because the wavelength is selected at intervals by the interleaver (first group branching filter). is there. Second, because the total number of cells in the LCOS when the bandwidth of a, 1 single channel and B _CH, has set FSR of the AWG, as follows.
FSR = 2 × a × B _CH formula (2)
Third, two different input / output ports are used to collect optical signals of different wavelength channels having a difference equivalent to 0.5 FSR through the spatial optical system including the AWG at the same condensing point on the LCOS. Provided, and by associating different input / output ports with each sub-wavelength group further multiplexed / demultiplexed by the second group demultiplexing filter (TFF).

  As already described, the feature of this embodiment is that the second group demultiplexing filter no longer requires steep filter characteristics. Referring back to FIG. 13B, for example, in TTF1, the optical signal of channel number 13 to be blocked with respect to the optical signal of channel number 9 to be transmitted has a sufficient wavelength (optical frequency) on the wavelength axis. ) Have an interval. Such a wavelength arrangement is obtained by combining an interleaver and a second group demultiplexing filter (TFF). As shown in FIG. 13B, in this embodiment, the filter characteristics of the TFF transmission band and stop band are shown as being adapted to the specific selected wavelengths of the wavelength groups G1 to G4, respectively. It was. TTF1 to TTF4 generally have filter characteristics that are shifted by one channel. In this embodiment, the filter characteristics of TTF can be designed gently. Therefore, the design and production of the TFF can be simplified by changing the filter characteristics by shifting the filter characteristics by one channel on the wavelength axis.

  FIG. 17 is a diagram illustrating a configuration of an AWG used in the variable dispersion compensator according to the fifth embodiment. FIG. 17 shows the configuration of only the AWG 901 in the reflective TODC block. Therefore, the condenser lens and the signal processing element (LCOS) are not displayed. The AWG 901 includes two input / output waveguides 902a and 902b, a slab waveguide 903, and an arrayed waveguide 904. The arrayed waveguide 904 is shown with the number of waveguides omitted. The two input / output waveguides are connected to the slab waveguide at the boundary surface at one end of the slab waveguide 903, but the connection point is separated by a distance that causes a difference in optical frequency corresponding to 0.5 FSR. Yes. The ends of the two input / output waveguides 902a and 902b correspond to port A and port B connected to the outside of the TODC block.

  FIG. 18 is a diagram illustrating another configuration of the AWG used in the variable dispersion compensator of the fifth embodiment. This is a configuration example in which a dielectric multilayer filter (TFF) 907 is inserted into a groove formed on an AWG chip. With this configuration, a single connection port with the interleaver can be provided. By forming a narrow groove in the PLC and inserting the dielectric multilayer filter 907 in this way, the dielectric multilayer filter can be mounted with only a slight excess loss. Using the configuration shown in FIG. 18, each TODC block in the dispersion compensator 600 of this embodiment shown in FIG. 12 can be mounted on one chip.

  In the first embodiment, the transmission band of one wavelength group and the AWG FSR are completely matched. However, in this embodiment, 22 channels are doubled for one wavelength group (11 channels). AWG with FSR equivalent to minutes is used. Further, the AWG has two input / output waveguides connected to two different ports, and the two input / output waveguides are connected to the slab waveguide at a position shifted by 0.5 FSR. Thereby, the influence of the light transmittance fall can be avoided. In this respect, the dispersion compensator of the present embodiment is common to the configuration of the second embodiment. Further, in the present embodiment, in one wavelength group that is group-demultiplexed, continuous adjacent points are obtained in that wavelengths (channels) are discretely and periodically selected by an interleaver (first group-demultiplexing filter). It should be noted that this is different from the first to fourth embodiments in which a plurality of channels are selected. A channel selected discretely and periodically according to a predetermined condition by an interleaver is further multiplexed and demultiplexed into two sub-wavelength groups by a second group demultiplexing filter, and corresponding to two different ports of the AWG. The filter characteristics of the second group demultiplexing filter (TFF) are relaxed to loose specifications. Furthermore, channels can be allocated to all cells on the LCOS without overlapping each other and without any gaps. In addition, it should be noted that the TODC block of this embodiment has the advantage that it can have the same configuration except that the TFF has different filter characteristics.

The configuration having the characteristics of the fifth embodiment can be generalized more widely as follows. Hereinafter, the relationship between the characteristics of the group demultiplexing filter, the number of wavelength groups demultiplexed by the group demultiplexing filter, and the number of AWG FSR and LCOS cells will be examined. As shown in FIG. 14, the number of wavelength groups demultiplexed by the group demultiplexing filter is b. In Example 5, b = 4. For simplicity, b is a power of 2 and b = 2, 4, 8, 16, 32,. The AWG FSR corresponds to FSR = 2 × a × B _CH (channel bandwidth) as shown in Equation (2), where a is the number of channels included in one wavelength group. Under the above conditions, the total number of channels (number of wavelengths) N processed by the dispersion compensator of this embodiment is N = a × b.

In the fifth embodiment, when a = 11 and the bandwidth B _CH of one channel is 100 GHz, the FSR is 2200 GHz. The number of LCOS cells is 11, and the number of channels is 11. Further, when a and b are in a prime relationship, channels can be allocated to all cells on the LCOS without overlapping each other and without a gap.

FIG. 19 is a diagram illustrating a relationship among the total number of wavelengths, the number of wavelength groups, and the number of LCOS elements applicable to the tunable dispersion compensator according to the fifth embodiment. The condition of this embodiment can be satisfied by selecting an appropriate number of LCOS element cells in accordance with the total number of channels (number of wavelengths) and the number of wavelength groups for which dispersion compensation is performed. Incidentally, the total number of channels N is necessary have names that are fully consistent with a × b.

  For brevity, b is assumed to be a power of 2. However, for example, even in the case of an interleaver having a filter characteristic that extracts a channel every 3 channels or every 5 channels, Note that similar ideas can be applied. In this case, the AWG's FSR is set to a predetermined relationship according to the specifications of the filter characteristics of the interleaver. Similarly, the AWG will have input / output waveguides that connect to the slab waveguides at different locations and will have a corresponding number of different ports. In this embodiment, since the number of LCOS cells used in one TODC block can be reduced and concentrated without any gaps, the LCOS element can be downsized. The number of cells actually created on the LCOS is not a problem, and even if the number of cells exceeds a, the specification of the filter characteristics of the second group demultiplexing filter (TFF) of this embodiment can be relaxed. It is only necessary that channels be assigned to LCOS cells without duplication or gaps.

  By using the tunable dispersion compensator of the present embodiment, as in the second embodiment, the number of channels in one wavelength group and the FSR are set to have a predetermined relationship, thereby avoiding the use of a region where the transmittance decreases in the transmission band of the FSR. , The influence of transmittance reduction can be avoided. A variable dispersion compensator with a flat light transmission band can be realized. It is also possible to relax the design conditions of the transmittance at the end of the transmission band of the AWG FSR.

  This embodiment can be applied to both a reflection type and a transmission type dispersion compensator. Further, the present invention is not limited to LCOS, and can also be applied to phase modulation elements including element elements. In addition, the fifth embodiment can be applied to an optical signal processing apparatus that performs signal processing other than the dispersion compensator in that a decrease in transmission characteristics of the AWG can be avoided.

  In the reflection type dispersion compensator shown in FIG. 12, the circulator 607 is used to separate the input signal to the dispersion compensator and the output signal. In the case of the dispersion compensator having the transmission type configuration of the first embodiment shown in FIG. 1, when each TODC block is a reflection type, a total of four, one for each of TODC1, TODC2, TODC3, and TODC4. A circulator is required. By providing the circulator 607 in front of the group demultiplexing filter 601 as in the present embodiment shown in FIG. 12, the number of circulators can be reduced.

  Also in this embodiment, phase characteristics having various shapes as described in the previous embodiments can be given to each cell.

  As described in detail above, a multi-channel variable dispersion compensator can be configured by further reducing the number of TODC blocks as compared with the variable dispersion compensator having the configuration of the first or second embodiment. Even when an interleaved group demultiplexing filter is used, the light transmission characteristics can be flattened. Further, by setting the FSR so as to have a predetermined relationship with the entire bandwidth of the communication channel group covered by the TODC block, it is possible to realize flattening of the light transmission characteristics.

  Also in the fifth embodiment, as described in detail in the third and fourth embodiments, the pixel is two-dimensionally configured using the 2D-LCOS element, so that the size in the direction corresponding to the spectral axis of the AWG is obtained. The same effect can be obtained as when using an LCOS element in which is kept short.

  The present invention can be used for a chromatic dispersion compensator in an optical communication system.

1, 34, 601, 901 AWG
3, 603, 903 Slab waveguide 4, 604, 904 Array waveguide 7, 36 Condensing lens 8, 37 Spatial phase modulation element 11, 14, 30, 607 Optical circulator 15 Bulk diffraction grating 16, 17, 902a, 902b Input / output waveguide 19, 401, 406, 601 Interleave type group demultiplexing filter 100, 200, 300, 600 Variable dispersion compensator 101, 102, 201, 202, 302, 303, 604a, 604b, 604c, 604d Group demultiplexing Filter 103, 103a, 103b, 103c, 103d, 203a, 203b, 203c, 203d, 400, 603a, 603b, 603c, 603d TODC block 301, 304 Demultiplexer 305, 407 2D-LCOS element 602 Optical fiber
907 TTF

Claims (6)

In a dispersion compensator for an interleaved communication system that performs processing by branching an optical signal group of consecutive N communication channels into a plurality of wavelength groups including channels discretely selected at predetermined channel intervals,
An interleaved group demultiplexing filter that multiplexes and demultiplexes the optical signal groups of the N communication channels into b first wavelength groups each having a maximum of a communication channels;
B dispersion compensator blocks corresponding to each of the b first wavelength groups,
Each dispersion compensator block is
A second group demultiplexing filter that further multiplexes and demultiplexes a corresponding one of the b first wavelength groups into two sub-wavelength groups, the b first wavelength groups A second group having a filter characteristic that transmits or blocks an optical signal for each of the a channels adjacently arranged with respect to the N communication channels. A demultiplexing filter ;
Optical multiplexing / demultiplexing means for further demultiplexing the optical signal group in the sub-wavelength group, and assuming that the bandwidth of one communication channel is B, it has an FSR of (2 × B × a), each of which is 1 / provided with two input-output ports having an optical frequency difference corresponding to 2FSR, an optical demultiplexing means for the input and output ports corresponding to each of the two sub wavelength group is selected,
A spatial phase modulation element including a plurality of element elements arranged in the direction of a demultiplexing axis of the optical multiplexing / demultiplexing means, wherein the plurality of element elements are divided into a sections each corresponding to one communication channel. A spatial phase modulation element ,
A dispersion compensator, wherein a and b are relatively prime and are selected so as to satisfy the relations of a ≧ 3, b ≧ 2 and N ≦ a × b.   The dispersion compensator according to claim 1, wherein the multiplexing / demultiplexing means is an arrayed waveguide diffraction grating (AWG), and the spatial phase modulation element is LCOS.   The dispersion compensator according to claim 1, wherein a power of 2 is selected for b. In each of the spatial phase modulation elements, a phase defined by a function of second order or higher is set independently for each section corresponding to each optical signal, with the distance of the demultiplexing axis as a parameter. The dispersion compensator according to claim 1. The spatial phase modulation element reflects an incident optical signal corresponding to the communication wavelength, and the optical group multiplexing / demultiplexing means and the optical multiplexing / demultiplexing means have a reflective configuration that multiplexes the dispersion-compensated optical signal. dispersion compensator according to claim 1 to 4, characterized in that.   6. The dispersion compensator according to claim 5, further comprising a circulator that inputs and outputs optical signal groups of the N communication channels on an input side of the interleaved group demultiplexing filter.