JP2013029758A - Core position specification method and alignment device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a core position specification method and an alignment device, which are capable of specifying positions of a plurality of cores in an end surface of a multi-core fiber in a shorter time.SOLUTION: In the core position specification method, a position in one end surface of a multi-core fiber, of one core included in N cores (N is an integer equal to or larger than 2) of the multi-core fiber is specified by rough alignment and fine alignment, and then, a light beam is made incident on predicted positions of (N-1) cores other than the one core on the assumption that the one core is each of N cores, on the basis of preliminarily acquired relative position information on N cores, and a core is determined as one of (N-1) cores if a light intensity in the predicted position of this core in the other end of the multi-core fiber exceeds a prescribed value, and positions of (N-1) cores are estimated in this manner. Each of estimated (N-1) cores is finely aligned, and the position of each of (N-1) cores is specified.

Description

  The present invention relates to a core position specifying method and an alignment apparatus.

  Patent Documents 1 and 2 describe a method of adjusting the optical axis of an optical module for coupling light emitted from a semiconductor laser to an optical fiber. In the method described in Patent Document 1, when adjusting the optical axes of a semiconductor laser unit having a semiconductor laser and a condenser lens and an optical fiber, the X and Y directions perpendicular to the optical axis and along the optical axis direction are adjusted. These are displaced relative to the Z direction, and their relative positions are matched so that the optical coupling efficiency is maximized. At this time, on the premise that the spatial power characteristic of the optical coupling in the plane perpendicular to the optical axis can be approximated to the Gaussian laser power distribution, the laser power at the optical axis position is based on the light intensity at four points in the same vertical plane. The amount of positional deviation from the focused beam diameter and the optical axis center is calculated. Further, in the method described in Patent Document 2, when the light emitting element and the optical fiber are relatively moved in a predetermined linear direction to align them, the light is emitted from the light emitting element and transmitted through the optical fiber. The light intensity is measured at a plurality of positions in the linear direction. Then, based on the measurement position and the light intensity measurement value, a correlation function between the relative position between the light emitting element and the optical fiber and the light intensity is created, and the light emitting element and the light are positioned at a position where the correlation function becomes a maximum value. Match the fiber. Non-Patent Document 1 describes an alignment algorithm for optically coupling an optical device and an optical fiber.

Japanese Patent Laid-Open No. 08-094886 Japanese Patent Application Laid-Open No. 08-094890

Toshiyuki Shima et al., “High-speed and high-precision optical axis alignment method for optical devices”, Matsushita Electric Engineering Technical Report, Vol. 55, no. 2

  In recent years, in order to cope with increasing optical communication traffic, multicore fibers having a plurality of cores have been actively studied. The multi-core fiber has, for example, a configuration in which a plurality of cores are two-dimensionally distributed and arranged in a cross section perpendicular to the longitudinal direction. However, existing alignment devices do not support multi-core fibers, and no suitable device or method for aligning optical axes between multi-core fibers or between multi-core fibers and other optical components yet exists. This is because, unlike the case of aligning a single core fiber in which one core is arranged at the center, when aligning a multi-core fiber, it is necessary to specify the position of each of the plurality of cores on the end face.

  Usually, when the position of the core on the end face of the optical fiber is specified, the end face of the other optical fiber is opposed to the end face of the optical fiber, and the other optical fiber is scanned over the entire end face. The position where the light intensity detected at the other end of the optical fiber increases is specified (coarse alignment). Thereafter, within a minute region including the position, while irradiating light from the other optical fiber, a position where the light intensity detected at the other end of the optical fiber is maximized is searched by a predetermined algorithm (fine alignment). ). In the above-mentioned Non-Patent Document 1, “raster alignment” and “spiral alignment” are described as scanning methods for coarse alignment, and “hill climbing alignment” is used as an algorithm for fine alignment. Is described.

  However, when the above-described core position specifying method is used for alignment of a multi-core fiber, coarse alignment and fine alignment are repeated each time a plurality of cores are specified. Therefore, it takes a long time to specify all the core positions.

  The present invention has been made in view of such problems, and provides a core position specifying method and an aligning device that can specify the positions of a plurality of cores on the end face of a multicore fiber in a shorter time. For the purpose.

  In order to solve the above-described problem, a core position specifying method according to the present invention is a method for specifying the positions of N cores on an end face of a multi-core fiber having N (N is an integer of 2 or more) cores. While observing one end face of the multi-core fiber, the attitude angle around the central axis of the one end face is fixed to a predetermined angle, and the intensity of light emitted from the N cores at the other end face of the multi-core fiber is integrated. A fiber fixing step for optically coupling the light detecting means to be detected to the other end face, and detecting the light intensity in the light detecting means while scanning the light beam two-dimensionally on the one end face, thereby including the N cores. Based on the first alignment step for specifying the position on the one end surface of the one core with the first accuracy and the relative position information of the N cores acquired in advance. Is assumed to be each of the N cores, a light beam is incident on the predicted position of the other (N-1) cores except for one core, and the light intensity in the light detection means exceeds a predetermined value. In this case, it is estimated by a core position estimating step for estimating the positions of (N-1) cores, assuming that one of (N-1) cores exists in the vicinity of the predicted position of the core, and the core position estimating step. By detecting the light intensity in the light detection means while scanning the light beam with respect to each of the (N-1) partial areas including the positions of the (N-1) cores, (N-1) pieces. And a second alignment step for specifying each position of the core with a second accuracy higher than the first accuracy.

In this core position specifying method, one end of a multi-core fiber is fixed in the fiber fixing step, and then the position of one core is specified first in the first alignment step. Thereafter, in the core position estimation step, other than the core when it is assumed that the core whose position is specified is each of the N cores based on the relative position information of the N cores acquired in advance. Predict the position of (N-1) cores. For example, in the case of having _three cores C _{1 to} C ₃ of a multi-core fiber, the core specified by the first alignment step is any one of the _three cores C _{1 to} C ₃ . Therefore, if the relative position information of the cores C _{1 to} C ₃ is acquired in advance, assuming that the core whose position is specified is the core C ₁ , the positions of the other cores C ₂ and C ₃ can be predicted. Similarly, assuming that the located core is core C ₂ , the positions of the other cores C ₁ and C ₃ can be predicted, and assuming that the located core is core C ₃ , the other cores It can predict the position of the C ₁ and _{C 2.}

  Light rays are incident on all the predicted core positions. When the light intensity in the light detection unit exceeds a predetermined value, one of the (N−1) cores exists in the vicinity of the predicted position of the core (that is, the core actually exists at the predicted position). (N-1) core positions can be estimated. Thus, after specifying (estimating) the positions of (N-1) cores with low accuracy, the positions of (N-1) cores are specified with high accuracy by the second alignment step.

  According to such a core position specifying method, after the position of one core is specified by the first alignment step, the position of another core can be specified without performing rough alignment. Therefore, since it is not necessary to repeat coarse alignment and fine alignment by the number of cores as in the conventional alignment method, the positions of a plurality of cores on the end face of the multicore fiber can be specified in a shorter time. In the present invention, the “neighboring” allows a deviation between the predicted position of the core and the center of the existing core, and when the light beam is incident on the predicted position of the core as described above, the existing core This includes a range in which it is possible to recognize that light has entered the light and that the light intensity in the light detection means has exceeded a predetermined value.

  In the core position specifying method, the light detection means detects the light intensity while scanning the light with respect to the partial region of the one end surface including the position of the one core specified in the first alignment step. It is preferable to further include a fine alignment step for specifying the position of one core with a third accuracy higher than the first accuracy. In this way, after the first alignment step, other (N−1) cores excluding the one core specified on the basis of the relative position information of the N cores acquired in advance. The prediction accuracy of the predicted position increases. Therefore, when a light beam is incident on the predicted position of the core, the light intensity in the light detection means when the core actually exists at the predicted position increases, and therefore the positions of (N−1) cores are easily estimated. can do.

  The core position specifying method is characterized in that a predicted position of at least one core included in (N-1) cores overlaps with a predicted position of other cores included in (N-1) cores. It is good. Thereby, since the number of predicted positions of the core can be reduced, the positions of a plurality of cores can be specified in a shorter time.

  Further, in the core position specifying method, N is an integer of 3 or more, and at least three cores included in the N cores are assumed to be regular M squares (M is an integer of 3 to N) assumed on one end face. It is good also as arrange | positioning in the position corresponded to each vertex. According to the research of the present inventor, in such a case, the predicted positions of (N-1) cores overlap each other, and the number of predicted positions of the core can be effectively reduced. Therefore, the positions of the plurality of cores can be specified in a shorter time.

  The alignment apparatus according to the present invention is an alignment apparatus that specifies the position of N cores on the end face of a multi-core fiber having N cores (N is an integer of 2 or more), and one end face of the multi-core fiber. A beam scanning means for two-dimensionally scanning a light beam, a light detection means for collectively detecting the light intensity of the light beam at the other end face of the multi-core fiber, and an observation means for observing one end face of the multi-core fiber; A support means for supporting one end face of the multi-core fiber so that a posture angle around the central axis of the one end face is variable; a storage means for storing relative position information of N cores on the one end face; And a calculation means for controlling the scanning means and specifying the positions of the N cores. The calculation means scans the light beam two-dimensionally with respect to the one end surface by the beam scanning means. By acquiring a detection signal from the light detection means, a position on one end face of one core included in the N cores is specified with the first accuracy, and based on relative position information stored in the storage means When one core is assumed to be each of the N cores, a light beam is made incident on the predicted positions of the other (N-1) cores excluding the one core, and the light intensity in the light detection means is predetermined. When the value is exceeded, the position of (N-1) cores is estimated by assuming that one of the (N-1) cores exists in the vicinity of the predicted position of the core, and the estimated (N-1) Each position of the (N-1) cores is obtained by acquiring a detection signal from the light detection means while scanning the light beam with respect to each of the (N-1) partial regions including the positions of the respective cores. Characterized by a second accuracy higher than the first accuracy To.

  In this alignment apparatus, the attitude angle around the central axis of the one end face can be fixed to a predetermined angle while observing the one end face of the multi-core fiber using the observation means and the support means. Furthermore, it is possible to collectively detect the intensity of light emitted from the N cores on the other end surface of the multi-core fiber by using the light detection means. And according to the calculation means of this alignment apparatus, after specifying the position of one core, the position of another core can be specified without performing rough alignment, as in the above-described core position specifying method. it can. Therefore, since it is not necessary to repeat the coarse adjustment core and the fine adjustment core by the number of cores, the positions of the plurality of cores on the end face of the multicore fiber can be specified in a shorter time.

  In the alignment apparatus, the calculation means detects from the light detection means while scanning the light beam by the beam scanning means with respect to the partial region of the one end surface including the position of the one core specified with the first accuracy. It is preferable to specify the position of one core with a third accuracy higher than the first accuracy by acquiring the signal. In this way, the (N-1) core positions other than the one core can be accurately estimated.

  The alignment apparatus is characterized in that a predicted position of at least one core included in (N-1) cores overlaps with a predicted position of other cores included in (N-1) cores. Also good. Thereby, since the number of predicted positions of the core can be reduced, the positions of a plurality of cores can be specified in a shorter time.

  Further, in the alignment apparatus, N is an integer of 3 or more, and at least three cores included in the N cores are assumed to be regular M squares (M is an integer of 3 to N) assumed on one end face. It may be arranged at a position corresponding to a vertex. According to the research of the present inventor, in such a case, the predicted positions of (N-1) cores overlap each other, and the number of predicted positions of the core can be effectively reduced. Therefore, the positions of the plurality of cores can be specified in a shorter time.

  In the alignment apparatus, it is preferable that the observation unit includes an imaging unit that images one end surface of the multi-core fiber. In this case, it is more preferable that the imaging unit is disposed away from the optical axis of the one end surface of the multicore fiber, and the observation unit further includes a reflecting mirror that reflects the image of the one end surface toward the imaging unit. Thus, the posture angle around the central axis of the one end surface can be adjusted to a predetermined angle while easily observing the one end surface of the multicore fiber.

  According to the core position specifying method and the alignment apparatus according to the present invention, the positions of a plurality of cores on the end face of the multicore fiber can be specified in a shorter time.

FIG. 1 is a diagram schematically showing an overall configuration of an alignment apparatus according to an embodiment of the present invention. FIG. 2 is a side view showing the alignment stage and its peripheral structure in detail. FIG. 3A is a side view showing a peripheral configuration in the vicinity of one end face of the multicore fiber and one end face of the single core fiber. FIG. 3B is a top view of the configuration shown in FIG. FIG. 4 is an enlarged perspective view showing an example of the configuration of the end of the optical fiber. FIG. 5 is a flowchart illustrating a core position specifying method according to an embodiment. FIG. 6 is a diagram showing how the angle of one end face of the multi-core fiber is adjusted, and shows a case where the number of cores is 7 as an example. FIG. 7 is a diagram illustrating an example of a light beam scanning method in the coarse alignment step. FIG. 8 is a diagram for explaining a method of searching for a position where the light intensity detected by the power sensor is maximum in the first fine alignment step. FIG. 9 is a diagram for explaining a core position prediction method. (A) has shown the actual core arrangement | positioning, (b) has shown the estimated position in the case of the core arrangement | positioning of (a). FIG. 10 is a diagram for explaining another example of the core position prediction method. (A) has shown the actual core arrangement | positioning, (b) has shown the estimated position in the case of the core arrangement | positioning of (a). FIG. 11 is a diagram for explaining another example of the core position prediction method. (A) has shown the actual core arrangement | positioning, (b) has shown the estimated position in the case of the core arrangement | positioning of (a). FIG. 12 is a diagram for explaining another example of the core position prediction method. (A) has shown the actual core arrangement | positioning, (b) has shown the estimated position in the case of the core arrangement | positioning of (a). FIG. 13 is a diagram for explaining another example of the core position prediction method. (A) has shown the actual core arrangement | positioning, (b) has shown the estimated position in the case of the core arrangement | positioning of (a). FIG. 14 is a diagram for explaining another example of the core position prediction method. (A) has shown the actual core arrangement | positioning, (b) has shown the estimated position in the case of the core arrangement | positioning of (a). FIG. 15 is a diagram for explaining another example of the core position prediction method. (A) has shown the actual core arrangement | positioning, (b) has shown the estimated position in the case of the core arrangement | positioning of (a). FIG. 16 is a diagram illustrating a predicted position when a core is disposed at each vertex and center of a regular hexagon as an example of a regular M-gon.

  Embodiments of a core position specifying method and an alignment apparatus according to the present invention will be described below in detail with reference to the accompanying drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

(Embodiment)
FIG. 1 is a diagram schematically showing an overall configuration of an alignment apparatus 1A according to an embodiment of the present invention. The alignment apparatus 1A according to the present embodiment has positions of all N cores on both end faces of the multi-core fiber 100 having N cores (N is an integer of 2 or more) and a clad covering the N cores. Can be identified. Each of the N cores of the multi-core fiber 100 preferably has a diameter (for example, 10 μm) for guiding light in a single mode. FIG. 1 shows an XYZ orthogonal coordinate system for easy understanding.

  As shown in FIG. 1, the alignment apparatus 1 </ b> A includes an alignment stage 20, a laser diode light source (hereinafter referred to as an LD light source) 11, an observation camera 13, and a display unit that displays an image of the observation camera 13 ( Monitor) 13a, a stage controller 15 for driving the alignment stage 20, and power sensors 16 and 17.

  The alignment device 1 </ b> A includes an arithmetic control unit 18 and a memory 19. The calculation control unit 18 is a calculation unit in the present embodiment. The arithmetic control unit 18 inputs signals from the power sensors 16 and 17, gives a control signal to the stage controller 15, and specifies the positions of the N cores on both end faces of the multicore fiber 100. The memory 19 is storage means in the present embodiment, and stores relative position information (core interval and relative core arrangement) of N cores on both end faces of the multi-core fiber 100.

  The alignment stage 20 includes a first side stage 21, a center stage 22, and a second side stage 23. These stages 21 to 23 are arranged in this order in a predetermined direction (Z-axis direction). The first side stage 21 and the second side stage 23 are supported so as to be movable with a degree of freedom of at least three axes. One end portion 100 a of the multicore fiber 100 is fixed to a region near the first side stage 21 on the central stage 22. The other end portion 100 b of the multicore fiber 100 is fixed to a region near the second side stage 23 on the center stage 22.

  In the region near the center stage 22 on the first side stage 21, one end 101a of the single core fiber 101 and the reflecting mirror 131 are fixed side by side. As the single core fiber 101, a single mode fiber is suitable. One end of the single core fiber 101 is disposed on one end surface side of the multicore fiber 100. One end 101a of the single core fiber 101 is supported by the first side stage 21 so as to be movable with at least three degrees of freedom. The other end 101 b of the single core fiber 101 is connected to the LD light source 11, whereby the other end of the single core fiber 101 and the LD light source 11 are optically coupled. The LD light source 11 and the single core fiber 101 constitute beam scanning means of the present embodiment, and scan the light beam two-dimensionally with respect to one end face of the multicore fiber 100. The LD light source 11 is a light source for alignment that generates laser light as a light beam, and generates infrared light, for example.

  One end portion 102 a of the single core fiber 102 is fixed to a region near the center stage 22 on the second side stage 23. As the single core fiber 102, a single mode fiber is suitable. The other end 102 b of the single core fiber 102 is connected to the power sensor 16, whereby the other end of the single core fiber 102 and the power sensor 16 are optically coupled. The single core fiber 102 and the power sensor 16 are used to detect the light intensity and the emission position of the light beam on the other end surface of the multicore fiber 100. That is, one end portion 102 a of the single core fiber 102 is disposed on the other end surface side of the multicore fiber 100 and is supported by the second side stage 23 so as to be movable with at least three degrees of freedom. The power sensor 16 is optically coupled to the other end surface of the single core fiber 102 and detects the intensity of light emitted from the other end surface.

  In the region near the center stage 22 on the second side stage 23, one end 103a of the large-diameter optical fiber 103 is fixed side by side with the one end 102a of the single core fiber 102. The other end 103 b of the large-diameter optical fiber 103 is connected to the power sensor 17, whereby the other end of the large-diameter optical fiber 103 and the power sensor 17 are optically coupled. The large-diameter optical fiber 103 and the power sensor 17 are light detection means for collectively detecting the intensity of light that is optically coupled to the other end face of the multi-core fiber 100 and emitted from the N cores.

  The observation camera 13 is an imaging unit in the present embodiment, and is for observing the relative position between the end face of the multicore fiber 100 and each end face of other optical fibers (single core fibers 101 and 102, large diameter optical fiber 103). Provided. The imaging data from the observation camera 13 is sent to the display unit 13a and the calculation control unit 18, and the first side stage 21 is controlled manually or automatically, so that one end surface of the multicore fiber 100 and the single core fiber 101 are controlled. One end faces each other. Alternatively, the second side stage 23 is controlled manually or automatically so that the other end surface of the multicore fiber 100 and one end surface of the single core fiber 102 or the large-diameter optical fiber 103 face each other.

  FIG. 2 is a side view showing in detail the alignment stage 20 and its peripheral structure of the present embodiment. As shown in FIG. 2, the alignment stage 20 includes at least a first drive unit 24 that supports the first side stage 21 movably with at least three degrees of freedom and a second side stage 23. And a second drive unit 25 that is movably supported with three or more degrees of freedom.

  The first drive unit 24 includes a Z stage 24a that translates the first side stage 21 in the Z-axis direction, an X stage 24b that translates the first side stage 21 in the X-axis direction, and a first side stage. A θz stage 24c that rotates 21 around the Z axis, a Y stage 24d that translates the first side stage 21 in the Y axis direction, and a θx stage 24e that rotates the first side stage 21 around the X axis. And a θy stage 24f that rotates and moves the first side stage 21 around the Y axis is connected on the base 26 in this order. The first side stage 21 is fixed to the θy stage 24f.

  Similarly, the second drive unit 25 includes a Z stage 25a that translates the second side stage 23 in the Z-axis direction, an X stage 25b that translates the second side stage 23 in the X-axis direction, A θz stage 25c that rotates the side stage 23 around the Z axis, a Y stage 25d that translates the second side stage 23 in the Y axis direction, and a second stage 23 that rotates around the X axis. A θx stage 25e and a θy stage 25f for rotating the second side stage 23 about the Y axis are connected on the base 26 in this order. The second side stage 23 is fixed to the θy stage 25f.

  The position of the central stage 22 is fixed on the base 26. Each stage of the first drive unit 24 and the second drive unit 25 described above is driven by a stage controller 15 (see FIG. 1).

  The observation camera 13 preferably includes an observation camera 13b disposed above the central stage 22 and an observation camera 13c disposed on the side of the central stage 22. Thereby, the relative positions of the multi-core fiber 100, the single-core fibers 101 and 102, and the large-diameter optical fiber 103 can be more accurately observed and controlled with high accuracy.

  Here, FIG. 3A is a side view showing a peripheral configuration in the vicinity of one end face of the multicore fiber 100 and one end face of the single core fiber 101. FIG. 3B shows a top view of the configuration shown in FIG. 3A, in which the observation camera 13b is represented by a virtual line. As shown in FIGS. 3A and 3B, the observation camera 13b is constituted by, for example, a CCD camera 13e and an objective lens 13f, and is arranged away from the optical axis L on one end face of the multicore fiber 100. Yes. The imaging surface of the CCD camera 13e faces the first side stage 21.

  On the first side stage 21, a reflecting mirror 131 is disposed in addition to the one end 101 a of the single core fiber 101 described above. The reflecting mirror 131 has a light reflecting surface inclined with respect to the optical axis of one end surface of the multicore fiber 100, and reflects an image of one end surface of the multicore fiber 100 toward the observation camera 13b. For the reflecting mirror 131, in addition to a member having a mirror surface, for example, a triangular prism or the like can be used. The observation camera 13b displays an image of one end face of the multicore fiber 100 on the display unit 13a. The reflecting mirror 131, together with the observation cameras 13b and 13c, constitutes observation means for observing one end surface of the multicore fiber 100. The reflecting mirror 131 is arranged on the first side stage 21 side by side with the one end 101a of the single core fiber 101 in the X-axis direction, and the first side stage 21 moves in the X-axis direction. Instead of the portion 101a, it faces one end surface of the multi-core fiber 100.

  4 shows one end portion 100a and the other end portion 100b of the multi-core fiber 100, one end portions 101a and 102a and the other end portions 101b and 102b of the single core fibers 101 and 102, and one end portion 103a and the other end of the large-diameter optical fiber 103. FIG. 4 is an enlarged perspective view showing an end portion 105a of an optical fiber 105 as an example of a configuration of a portion 103b. The end portion 105 a is formed by attaching a fiber holding member 106 to the optical fiber 105. The fiber holding member 106 is a glass capillary having an outer diameter of about 2.5 mm, for example, and the end face thereof is polished after the optical fiber 105 is mounted. When the fiber holding member 106 is such a glass capillary, since the internal optical fiber 105 can be visually recognized when observing with the observation camera 13, it is easy to match the relative positions of the optical fibers. The fiber holding member 106 may be a zirconia ferrule, or may be a square member used for a single core fiber array.

  Subsequently, a core position specifying method using the alignment device 1A described above will be described. FIG. 5 is a flowchart illustrating a method according to the present embodiment.

<Fiber fixing step>
First, the other end portion 100b of the multi-core fiber 100 and the one end portion 103a of the large-diameter optical fiber 103 are brought close to each other, and these end faces are opposed to each other to be optically coupled (step S11 in FIG. 5). That is, the observation camera 13 is moved above (and to the side of) the other end portion 100b, and the other end portion 100b and the one end portion 103a are moved while confirming the relative positions of the optical fibers 100 and 103 on the display portion 13a. Move closer together.

  Subsequently, the one end portion 100a of the multi-core fiber 100 and the one end portion 101a of the single core fiber 101 are brought close to each other, and these end faces are opposed to each other (step S12 in FIG. 5). That is, the observation camera 13 is moved above (and to the side) the one end portions 100a and 101a, and the one end portions 100a and 101a are moved while confirming the relative positions of the one end faces of the optical fibers 100 and 101 on the display portion 13a. Move closer together.

  In steps S11 and S12 described above, the arithmetic control unit 18 shown in FIG. 1 performs image processing and the like, and grasps the relative positions of the one end portions 100a and 101a (or the other end portion 100b and the one end portion 103a). The calculation control unit 18 may send the control signal to the stage controller 15 so that the one end portions 100a and 101a (or the other end portion 100b and the one end portion 103a) are brought close to each other.

  Subsequently, by moving the first side stage 21 by a predetermined distance in the X-axis direction, the reflecting mirror 131 on the first side stage 21 is arranged on the optical axis of the one end portion 100a of the multicore fiber 100, and reflected. The light reflecting surface of the mirror 131 and the one end surface of the multicore fiber 100 are opposed to each other. Then, while observing an image of one end surface of the multi-core fiber 100 incident on the observation camera 13b through the reflecting mirror 131, the posture angle around the central axis of the one end surface of the multi-core fiber 100 is made closer to a predetermined angle (step in FIG. 5). S13).

  Here, FIG. 6 is a diagram showing a state of angle adjustment of one end face of the multi-core fiber 100, and shows a case where the number of cores is 7 (ie, N = 7) as an example. Normally, as shown in FIG. 6A, at the stage where the one end portion 100a of the multicore fiber 100 is installed on the central stage 22, the posture angle around the central axis of one end surface of the multicore fiber 100 (in the figure, the multicore fiber is shown). The angle θ formed by the reference line A of 100 posture angles and the X axis is indefinite. Therefore, in this embodiment, the posture angle θ is brought close to a predetermined angle (0 ° in the figure) while observing an image of one end face of the multicore fiber 100 using the reflecting mirror 131 and the observation camera 13b. As a result, the N cores 100c on one end surface of the multi-core fiber 100 are arranged at predetermined coordinate positions on the XY coordinate plane. After the posture angle θ is adjusted to a predetermined angle in this way, the one end portion 100 a of the multicore fiber 100 is fixed to the central stage 22.

  In order to suitably perform step S13 described above, the alignment device 1A of the present embodiment supports the one end surface of the multicore fiber 100 so that the posture angle θ around the central axis of the one end surface is variable. Supporting means such as a rotating fiber holder is further provided. As such a rotating fiber holder, for example, F264N manufactured by Suruga Seiki is suitably used.

<First alignment step>
Subsequently, in a state where the one end surface of the single core fiber 101 is brought close to the one end surface of the multicore fiber 100, the arithmetic control unit 18 transmits the light beam emitted from the single core fiber 101 to the one end surface of the multicore fiber 100. To scan in two dimensions. At this time, the size of the scanned region is, for example, several hundred micrometers square. In order to increase the optical coupling efficiency between one end face of the multi-core fiber 100 and one end face of the single core fiber 101, it is preferable to apply a refractive index matching oil between these end faces. During this scanning, the light sensor 17 detects the light intensity of the light beam on the other end surface of the multi-core fiber 100 through the large-diameter optical fiber 103. At this time, when the light beam emitted from the single core fiber 101 passes near each core of the multicore fiber 100, the detection signal output from the power sensor 17 increases. Therefore, for example, when the detection signal from the power sensor 17 exceeds a predetermined threshold, the position of the single core fiber 101 at that time can be determined as the position of a certain core. Thus, on one end face of the multi-core fiber 100, the approximate position of one core (hereinafter referred to as a specific core) included in the N cores is specified with a slightly coarse first accuracy (in the example, about 5 μm) ( Step S14 in FIG. This step is completed when the position of one specific core is specified.

  Here, FIG. 7 is a diagram illustrating an example of a light beam scanning method in the first alignment step S14 described above. As shown in FIG. 7A, in one example, the multi-core fiber 100 is gradually moved in a direction along one end face and moved in the opposite direction while gradually moving in a direction crossing the direction. A method of moving the fiber 101 (so-called raster alignment) is suitable. Further, as shown in FIG. 7B, in another example, scanning is started from the vicinity of the center of one end face of the multicore fiber 100, and the single core fiber 101 is moved in a spiral shape (so-called spiral alignment). Is also suitable. The interval between the reciprocating movement lines in the raster alignment or the interval between the spiral movement lines in the spiral alignment is, for example, 5 μm.

<Fine adjustment step>
Subsequently, the calculation control unit 18 scans the light beam emitted from the single core fiber 101 in a state where one end face of the single core fiber 101 is brought close to a partial region including the position of the specific core. During this scanning, the light sensor 17 detects the light intensity of the light beam on the other end surface of the multi-core fiber 100 via the large-diameter optical fiber 103. In addition, the arithmetic control unit 18 searches for a position where the light intensity becomes maximum by moving the single core fiber 101 using a predetermined algorithm in a direction in which the signal detected by the power sensor 17 becomes larger. In this way, the detailed position of the specific core is specified with the third accuracy higher than the first accuracy on one end face of the multi-core fiber 100 (step S15 in FIG. 5). The third accuracy is, for example, on the order of submicron (in the example, 0.1 μm).

  Here, FIG. 8 is a diagram for explaining a method (mountain climbing method) for searching for a position where the light intensity detected by the power sensor 17 is maximum in the fine alignment step S15 described above. As shown in FIG. 8A, first, the single core fiber 101 is moved along a certain linear direction A1. Then, after moving by a certain length, the position having the highest light intensity within the moved range is specified, and the single core fiber 101 is moved again to that position. Next, starting from that position, the single core fiber 101 is moved along a direction A2 orthogonal to the direction A1, and after moving by a certain length, a position having the highest light intensity is specified within the moved range. . By repeating such an operation (FIG. 8B), a position where the light intensity is maximum can be searched.

<Core position estimation step>
Subsequently, the arithmetic control unit 18 estimates the positions of (N−1) cores excluding the specific core. In this step, first, when the arithmetic control unit 18 assumes that the specific core is each of the N cores based on the relative position information of the N cores stored in the memory 19, The positions of (N-1) cores are predicted (step S16 in FIG. 5).

FIG. 9 is a diagram for explaining such a core position prediction method. FIG. 9A shows seven cores C _{1 to} C ₇ as an example of the core arrangement on one end face of the multi-core fiber 100. The six cores C _{2 to} C ₇ included in the _seven cores C _{1 to} C ₇ are arranged at positions corresponding to the vertices of the regular hexagon H 1 assumed on one end face of the multi-core fiber 100. Also, one of the core C ₁ is arranged in the center of the regular hexagon H1. FIG. 9B shows the core (specific core) C _S whose position is specified by the first alignment step S14 and the fine alignment step S15 described above when the multi-core fiber 100 has such a core arrangement. FIG. 7 is a diagram showing positions (hereinafter, referred to as predicted positions) C _{P1 to} C _P18 related to other six cores predicted from _A.

In a first aligning step S14 and the fine alignment step S15 described above, of the seven core _C 1 -C ₇ shown in FIG. 9 (a), the position of one particular core _{C S} is specified. Assuming that this specific core _CS is, for example, the core C ₁ in FIG. 9A, the predicted positions of the other six cores are the predicted positions C _{P1 to} C _P6 shown in FIG. 9B. Become. Also, assuming a specific core _{C S,} for example a core _{C 2} in FIG. 9 (a), the predicted location of the other six cores predicted position shown in FIG. _{9 (b) C P1, C} P2 , C _{P6 to} C _P9 . Also, assuming a specific core _{C S} and the core _{C 3,} for example FIG. 9 (a), the predicted location of the other six cores predicted position shown in FIG. 9 (b) _C P1 ~C _P3 , C _{P9 to} C _P11 . As described above, assuming that the specific core C _S is each of the _seven cores C _{1 to} C ₇ , the predicted positions C _P1 to C shown in FIG. 9B are predicted positions of the other six cores. _CP18 can be determined.

Determination of the position coordinates of the predicted positions C _{P1 to} C _P18 can be performed as follows, for example. First, a specific core _{C S} and the coordinate origin, a vector representing the predicted position _{C P1} a, b a vector representing the predicted position _{C P2,} a vector representing the predicted position _{C P3} and c. When these cores a to c are expressed in the XY coordinate system of the first side stage 21 with the core interval of 30 μm,
a = (30 × cos 60 °, 30 × sin 60 °)
b = (30 × cos0 °, 30 × sin0 °)
c = (30 × cos300 °, 30 × sin300 °)
It becomes. Then, after expressing the predicted positions C _{P1 to} C _P18 with these vectors, XY coordinate conversion may be performed. The vector notation and XY coordinates of the predicted positions C _{P1 to} C _P18 are as shown in Table 1 below.

In the core position estimation step, subsequently, the calculation control unit 18 sequentially moves the one end 101a of the single core fiber 101 to the predicted positions C _{P1 to} C _P18 obtained as described above, and the predicted positions C _{P1 to} C C _A light beam is incident on each of _P18 . The light intensity of the light beam on the other end surface of the multicore fiber 100 is detected by the power sensor 17 via the large-diameter optical fiber 103. At this time, for example, when the detection signal from the power sensor 17 exceeds a predetermined threshold (for example, −15 dB with respect to the intensity of light emitted from the LD light source 11), the position of the single core fiber 101 at that time is It is determined that the position is _{one of the} cores C _{1 to} C ₇ (that is, the core exists in the vicinity of the predicted position). In this way, the approximate positions of the other six cores excluding the specific core can be specified with slightly coarse accuracy, and the positions of these cores can be estimated (step S17 in FIG. 5). The accuracy in this step is, for example, approximately the same as that in the first alignment step S14.

On the other hand, in the present embodiment, “neighboring” allows a deviation between the predicted position of the core and the center of the existing core, and exists when a light beam is incident on the predicted position of the core as described above. This includes a range in which light is incident on the core and the light intensity in the light detection means can be recognized as exceeding a predetermined value. That is, in the present embodiment, even if the center of the one end portion 101a of the single core fiber 101 is deviated from the center position of any of the cores C _{1 to} C _{7 by} , for example, about 10 μm, A detection signal can be detected. Note that this error range may vary depending on the detection sensitivity of the power sensor 17 and the set value of the threshold.

If the power sensor 17 has sufficient detection sensitivity, the fine alignment step may be omitted. In this case, since the alignment accuracy in one core C ₁ is lowered, the position accuracy of the predicted positions C _{P1 to} C _P6 is also lowered, so that the single core fiber 101 and the multi-core fiber 100 are A loss will occur at the joint. Therefore, by setting the predetermined threshold value low, the fine alignment step can be omitted and the core position can be estimated more quickly. However, in order to more clearly distinguish the position where the core exists from the position where the core does not exist and accurately estimate the core position, it is more preferable to perform the fine alignment step.

<Second alignment step>
Subsequently, the calculation control unit 18 sequentially brings one end face of the single core fiber 101 close to each of the six partial regions including the positions of the six cores estimated by the core position estimation step, The light beam emitted from the core fiber 101 is scanned. During this scanning, the light sensor 17 detects the light intensity of the light beam on the other end surface of the multi-core fiber 100 via the large-diameter optical fiber 103. In this step, the detailed positions of the six cores are set with the second accuracy higher than the first accuracy (typically equal to the third accuracy) by the same method as in the fine alignment step S15 described above. Therefore, it specifies (step S18 of FIG. 5).

<Other end face alignment step>
Subsequently, the arithmetic control unit 18 moves the second side stage 23 in the X-axis direction, thereby bringing the other end portion 100b of the multicore fiber 100 and the one end portion 102a of the single core fiber 102 closer to each other, and these end faces. Face each other. That is, the observation camera 13 is moved above (and to the side of) the other end portion 100b, and the other end portion 100b and the one end portion 102a are moved while confirming the relative positions of the optical fibers 100 and 102 on the display portion 13a. Move closer together.

Then, at one end 100 a of the multicore fiber 100, the one end 101 a of the single core fiber 101 is optically coupled to one of the _seven cores C _{1 to} C ₇ , and then at the other end 100 b of the multicore fiber 100. Then, one end 102a of the single core fiber 102 is scanned to perform coarse alignment and fine alignment (step S19 in FIG. 5). Thereby, the one end part 102a of the single core fiber 102 and the said core can be optically coupled in the other end surface of the multi-core fiber 100. Note that after this step, the multi-core fiber 100 and the single-core fibers 101 and 102 may be fixed to each other with an adhesive as necessary.

  According to the alignment apparatus 1A and the core position specifying method of the present embodiment described above, after the position of one core is specified by the first alignment step S14 and the fine alignment step S15, the coarse alignment step is performed. The position of another core can be specified without performing. Therefore, since it is not necessary to repeat the coarse alignment step and the fine alignment step as many as the number of cores as in the conventional alignment method, the positions of a plurality of cores on the end face of the multicore fiber can be specified in a shorter time. .

(Modification)
In the above embodiment, the example in which the _seven cores C _{1 to} C ₇ are arranged at each vertex and center of the regular hexagon H 1 as shown in FIG. 9A has been described. The apparatus 1A and the core position specifying method are not limited to such a core arrangement and can be applied to various core arrangements. 10-15 is a figure for demonstrating the example of various core arrangement | positioning. 10 to 15, (a) shows an actual core arrangement, and (b) shows a predicted position in the case of the core arrangement of (a).

FIG. 10A shows an example in which _four cores C _{1 to} C ₄ are arranged at each vertex and center of the equilateral triangle T. Further, FIG. 10 (b), in the case of such a core arrangement, the other three cores predicted from the specific core C _S whose position is identified by the first aligning step S14 and the fine alignment step S15 Predicted positions C _{P1 to} C _P12 are shown.

For example, assuming that the specific core C _S is the core C ₁ in FIG. 10A, the predicted positions of the other three cores are the predicted positions C _P1 , C _P2 , _CP3 . Also, assuming a specific core _{C S} and the core _{C 2,} for example FIG. 10 (a), the predicted location of the other three core predicted position shown in FIG. _{10 (b) C P4, C} P5 , _CP6 . Also, assuming a specific core _{C S} and the core _{C 3,} for example FIG. 10 (a), the predicted location of the other three cores are predicted position _C P7 shown in FIG. 10 _{(b), C P8} , _CP9 . As described above, assuming that the specific core C _S is each of the _four cores C _{1 to} C ₄ , the predicted positions C _P1 to C shown in FIG. 10B as the predicted positions of the other three cores _CP12 can be determined.

FIG. 11A shows an example in which _five cores C _{1 to} C ₅ are arranged at each vertex and center of the regular square S. Further, FIG. 11 (b), in the case of such a core arrangement, other four cores predicted from the specific core C _S whose position is identified by the first aligning step S14 and the fine alignment step S15 Predicted positions C _{P1 to} C _P12 are shown.

For example, assuming that the specific core _CS is the core C ₁ in FIG. 11A, the predicted positions of the other three cores are the predicted positions C _{P1 to} C _P4 shown in FIG. Become. Also, assuming a specific core _{C S,} for example a core _{C 2} in FIG. 11 (a), the predicted location of the other four core predicted position shown in FIG. _{11 (b) C P1, C} P5 ~ _CP7 . Also, assuming a specific core _{C S} and the core _{C 3,} for example FIG. 11 (a), the predicted location of the other four core predicted position shown in FIG. _{11 (b) C P2, C} P5 , C _P8 and C _P9 . As described above, assuming that the specific core C _S is each of the _five cores C _{1 to} C ₅ , the predicted positions C _P1 to C shown in FIG. 11B as the predicted positions of the other four cores _CP12 can be determined.

FIG. 12A shows an example in which _six cores C _{1 to} C ₆ are arranged at each vertex and center of a regular pentagon P. Further, FIG. 12 (b), in the case of such a core arrangement, other five cores predicted from the specific core C _S whose position is identified by the first aligning step S14 and the fine alignment step S15 30 predicted positions _CP are shown.

FIG. 13A shows an example in which _nine cores C _{1 to} C ₉ are arranged at each vertex and center of a regular octagonal OC. Further, FIG. 13 (b), in the case of such a core arrangement, other eight cores predicted from the specific core C _S whose position is identified by the first aligning step S14 and the fine alignment step S15 40 predicted positions _CP are shown.

FIG. 14A shows the seven cores C _{1 to} C ₇ arranged at the vertices and the center of the regular hexagon H2, and the center position and the direction from the center position to the vertices with respect to the regular hexagon H2. 12 cores C _{8 to} C ₁₉ arranged between each vertex of adjacent regular hexagon H3 and adjacent vertices are shown. Further, FIG. 14 (b), in the case of such a core arrangement, other 18 cores predicted from the specific core C _S whose position is identified by the first aligning step S14 and the fine alignment step S15 60 predicted positions _CP are shown.

FIG. 15A shows an example in which _four cores C _{1 to} C ₄ are arranged on a straight line L1. Further, FIG. 15 (b), in the case of such a core arrangement, the other three cores predicted from the specific core C _S whose position is identified by the first aligning step S14 and the fine alignment step S15 Predicted positions C _{P1 to} C _P6 are shown.

For example, assuming that the specific core _CS is the core C ₁ in FIG. 15A, the predicted positions of the other three cores are the predicted positions C _{P1 to} C _P3 shown in FIG. Become. Also, assuming a specific core _{C S} and the core _{C 2,} for example FIG. 15 (a), the predicted location of the other three cores, 15 predicted position shown in (b) _C P2 ~C _P4 It becomes. Also, assuming a specific core _{C S} and the core _{C 3,} for example FIG. 15 (a), the predicted location of the other three cores, 15 predicted position shown in (b) _C P3 ~C _P5 It becomes. As described above, assuming that the specific core C _S is each of the _four cores C _{1 to} C ₄ , the predicted positions CP ₁ to C shown in FIG. 15B as the predicted positions of the other three cores _CP6 can be determined.

In the core position specifying method described above, the predicted position of at least one core included in (N-1) cores excluding the specific core is the same as the predicted position of other cores included in (N-1) cores. It is preferable to overlap. For example, in the example illustrated in FIG. 9, it is assumed that the specific position C _P2 corresponding to the core C ₁ when the specific core C _S is the core C ₃ and the specific core C _S is the core C _2. a predicted position C _P2 corresponding to the core C ₇ in the case of overlap with each other (match). As described above, when the predicted positions of two or more cores overlap (match) each other, the number of predicted positions of the cores can be reduced. Therefore, since the time required for estimating the core position can be shortened, the positions of a plurality of cores can be specified in a shorter time.

  Moreover, in the core position specifying method described above, an example in which the core is arranged at each vertex of a regular polygon (regular triangle, regular square, regular pentagon, regular hexagon, and regular octagon) is shown in FIGS. . In this way, at least three cores included in N (N is an integer of 3 or more) cores correspond to the vertices of a regular M-gon (M is an integer of 3 or more and N or less) assumed on one end face. It is preferable to arrange in the position. According to the research of the present inventor, in such a case, among the predicted positions of (N-1) cores excluding a specific core, many predicted positions overlap each other, and the number of predicted positions of the core is effectively reduced. be able to. Therefore, it is possible to specify the positions of the plurality of cores in a shorter time.

  Such overlapping of predicted positions will be further examined. When cores are arranged at each vertex and center of a regular M-gon, the number of cores is (M + 1). If the case where (M + 1) cores are irregularly arranged is considered, there are M prediction positions for one core, so there are {M × (M + 1)} prediction positions in total. It will be. On the other hand, for example, as shown in FIG. 11 (b), when the core is arranged at each vertex and center of a regular square, there should be 4 × (4 + 1) = 20 predicted positions. However, since the predicted positions overlap each other, the actual number of predicted positions is twelve. For example, as shown in FIG. 9B, when a core is arranged at each vertex and center of a regular hexagon, originally there should be 6 × (6 + 1) = 42 predicted positions. The actual number of predicted positions is 18.

FIG. 16 is a diagram (see FIG. 9B) showing predicted positions C _{P1 to} C _P18 when a core is arranged at each vertex and center of a regular hexagon as an example of a regular M-gon. As described above, when the cores are arranged at the vertices and the centers of the regular M-polygon, the predicted positions overlap when different cores are used as the specific cores. For example, in FIG. 16, one vertex of the predicted regular hexagon H4 and one vertex of another regular hexagon H5 overlap each other at the mark B1. Further, when there is an outer peripheral core disposed at a point-symmetrical position with respect to the central core, the central core and the outer peripheral core overlap each other (mark B2 in the figure). Furthermore, if there is another predicted position at a position that is a line perpendicular to the specific core and an arbitrary predicted position and is symmetric with respect to the straight line passing through the specific core, a regular hexagon centered on the specific core The predicted position at one vertex and the predicted position at one vertex of another regular hexagon (for example, H4) overlap each other. If any of these prediction positions overlaps, the number of predicted positions can be reduced. In particular, when the core is arranged at each vertex and center of the regular hexagon as illustrated in FIG. 16, all of these overlapping modes exist, and therefore, the overlapping of the predicted positions becomes extremely large. This is the most preferable form.

  The core position specifying method and the alignment apparatus according to the present invention are not limited to the above-described embodiments, and various other modifications are possible. For example, in the above-described embodiment, the “mountain climbing method” is exemplified as a method for performing the fine alignment step and the second alignment step. However, the method for performing these steps is not limited thereto, and the first alignment step is not limited thereto. Various methods can be used as long as the accuracy can be higher than the accuracy in the lead step.

  The multi-core fiber in the present invention is not limited to the above-described embodiment, and may be a multi-core holey fiber in which a large number of holes are arranged so as to surround the core region, for example. That is, the clad of the multi-core fiber in the present invention is not particularly limited as long as it has an action of surrounding the core region and confining the transmission light in the core region.

DESCRIPTION OF SYMBOLS 1A ... Alignment apparatus, 11 ... Light source, 13 ... Observation camera, 13a ... Display part, 15 ... Stage controller, 16, 17 ... Power sensor, 18 ... Calculation control part, 19 ... Memory, 20 ... Alignment stage, 21 ... 1st side stage, 22 ... center stage, 23 ... 2nd side stage, 24 ... 1st drive part, 25 ... 2nd drive part, 26 ... base, 100 ... multi-core fiber, 101,102 ... single core fiber , 103 ... large diameter optical fiber, 105 ... optical fiber, 106 ... fiber holding member, 131 ... _reflector, C 1 -C _{19 ... core, C P, C P1 ~C P18} ... predicted position, C S _... particular core.

Claims (10)

A method for identifying the position of the N cores on an end face of a multi-core fiber having N cores (N is an integer of 2 or more),
While observing one end face of the multi-core fiber, the posture angle around the central axis of the one end face is fixed to a predetermined angle, and the intensity of light emitted from the N cores on the other end face of the multi-core fiber is adjusted. A fiber fixing step for optically coupling light detecting means for collectively detecting to the other end surface;
The light detection means detects the light intensity while two-dimensionally scanning the light with respect to the one end surface, thereby determining the position of the one core included in the N cores on the one end surface with a first accuracy. A first alignment step identified with
Other (N-1) except for the one core when it is assumed that the one core is each of the N cores based on the relative position information of the N cores acquired in advance. When a light beam is incident on a predicted position of each core and the light intensity in the light detection unit exceeds a predetermined value, one of the (N-1) cores is present in the vicinity of the predicted position of the core. A core position estimating step of estimating the positions of the (N-1) cores;
The light detection means detects the light intensity while scanning the light beam for each of the (N-1) partial areas including the positions of the (N-1) cores estimated by the core position estimation step. And a second alignment step for specifying each position of the (N-1) cores with a second accuracy higher than the first accuracy. Method.   The light detection means detects the light intensity while scanning the partial area of the one end surface including the position of the one core specified in the first alignment step, thereby the one core. The core position specifying method according to claim 1, further comprising a fine alignment step of specifying the position of the first position with a third accuracy higher than the first accuracy.   The predicted position of at least one core included in the (N-1) cores overlaps with a predicted position of other cores included in the (N-1) cores. Or the core position specifying method according to 2.   N is an integer of 3 or more, and at least three cores included in the N cores correspond to the vertices of a regular M-gon (M is an integer of 3 to N) assumed on the one end face. The core position specifying method according to any one of claims 1 to 3, wherein the core position specifying method is provided in any one of claims 1 to 3. An alignment device that identifies the position of the N cores on the end face of a multi-core fiber having N cores (N is an integer of 2 or more),
Beam scanning means for scanning a light beam two-dimensionally with respect to one end face of the multi-core fiber;
A light detecting means for collectively detecting the light intensity of the light beam at the other end surface of the multi-core fiber;
An observation means for observing the one end face of the multi-core fiber;
Supporting means for supporting the one end face of the multi-core fiber so that a posture angle around a central axis of the one end face is variable;
Storage means for storing relative position information of the N cores on the one end surface;
And controlling the beam scanning means, and comprising calculating means for specifying the positions of the N cores,
The computing means is
By acquiring a detection signal from the light detection means while scanning the light beam two-dimensionally with respect to the one end face by the beam scanning means, the one end face of one core included in the N cores Locating with first accuracy,
Based on the relative position information stored in the storage unit, other (N−1) cores excluding the one core when the one core is assumed to be each of the N cores It is assumed that one of the (N-1) cores exists in the vicinity of the predicted position of the core when the light beam is incident on the predicted position and the light intensity in the light detection unit exceeds a predetermined value. -1) estimating the position of one core,
By acquiring a detection signal from the light detection means while scanning a light beam with respect to each of the (N-1) partial regions including the estimated positions of the (N-1) cores, An alignment apparatus, wherein each position of (N-1) cores is specified with a second accuracy higher than the first accuracy.   The calculation means detects a detection signal from the light detection means while causing the beam scanning means to scan a partial area of the one end face including the position of the one core specified with the first accuracy. The alignment apparatus according to claim 5, wherein the position of the one core is specified with a third accuracy higher than the first accuracy by acquiring.   The predicted position of at least one core included in the (N-1) cores overlaps with a predicted position of other cores included in the (N-1) cores. Or the alignment apparatus according to 6;   N is an integer of 3 or more, and at least three cores included in the N cores correspond to the vertices of a regular M-gon (M is an integer of 3 to N) assumed on the one end face. The alignment device according to any one of claims 5 to 7, wherein the alignment device is disposed in the center.   The alignment device according to any one of claims 5 to 8, wherein the observation unit includes an imaging unit that images the one end surface of the multi-core fiber. The imaging means is disposed away from the optical axis of the one end face of the multi-core fiber;
The alignment apparatus according to claim 9, wherein the observation unit further includes a reflecting mirror that reflects the image of the one end surface toward the imaging unit.

Images