Azimuthal alignment of anti-resonant hollow core fibers

Abstract

A method of azimuthally aligning anti-resonant hollow fibers (ARFs) in a fusion splicer is described. The method includes, for each anti-resonant hollow fiber: inserting the anti-resonant hollow fiber into the fusion splicer, laterally illuminating the anti-resonant hollow fiber, capturing intensity data for each pixel in a row of pixels in an image of the laterally illuminated anti-resonant hollow fiber and processing the intensity data to generate processed intensity data; and determining, from the processed intensity data, a pixel position in the row corresponding to a predefined feature (e.g., a maximum intensity value) of the processed intensity data, the pixel position corresponding to an initial azimuthal orientation. The method further includes rotating at least one of the two anti-resonant hollow fibers based on the determined pixel position of the predefined feature.

Description

Background Technology

[0001] Fusion splicing is commonly used to connect optical fibers, and several splicing methods can be used, including fusion splicing and mechanical splicing. Fusion splicing involves aligning the fiber ends to each other and securing them in the aligned position. For example, fusion splicing connects two optical fibers by heating the end regions to soften the glass used to manufacture the fiber. By pressing the ends together, the softened glass is fused, thus permanently connecting the fibers as the glass cools and hardens. In contrast, mechanical splicing does not permanently connect the optical fibers together; instead, it uses mechanical means to maintain their alignment and secure the fiber ends together.

[0002] A connection formed by fusion splicing two optical fibers together is called a fusion splice. The quality of the fusion splice is a crucial factor in achieving low-loss light propagation when light travels from one fiber to another. Accurately aligning the structural features within the two fibers to reduce structural discontinuities in the fusion splice contributes to achieving low loss.

[0003] Traditional solid-core optical fibers, including a ring-shaped cladding surrounding a circular core, are relatively easy to align and splice. These structures possess continuous rotational symmetry in their cross-section; therefore, by aligning the fiber ends laterally to match the position of the fiber's longitudinal axis, alignment of the core and cladding is guaranteed. However, antiresonant hollow-core optical fibers have a complex internal structure and lack continuous rotational symmetry. Misalignment of this internal structure increases optical loss as light travels from one fiber to another at the splice point.

[0004] The embodiments described below are not limited to implementations that address any or all of the drawbacks of known fiber locations or fiber fusion splicing devices. Summary of the Invention

[0005] The following is a simplified summary of the invention to provide the reader with a basic understanding. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to limit the scope of the claimed subject matter. Its sole purpose is to present selected concepts disclosed herein in a simplified form as an introduction to the more detailed description that follows.

[0006] A method for azimuth alignment of anti-resonant hollow fiber (ARF) in a fusion splicer is described. The method includes, for each ARF: inserting the ARF into the fusion splicer; side-illuminating the ARF; capturing intensity data for each pixel in a pixel row of a side-illuminated ARF image and processing the intensity data to generate processed intensity data; and determining, based on the processed intensity data, a pixel position in the row corresponding to a predefined feature (e.g., a maximum intensity value) of the processed intensity data, the pixel position corresponding to an initial azimuth direction. The method also includes rotating at least one of the two ARFs based on the determined pixel position of the predefined feature.

[0007] Many of the accompanying features will be more readily understood by referring to the following specific embodiments considered in conjunction with the accompanying drawings. Attached Figure Description

[0008] This specification will be better understood by referring to the following detailed embodiments in conjunction with the accompanying drawings, wherein: Figures 1 to 3 Cross-sectional views of three different examples of anti-resonant hollow optical fibers are shown. Figure 4 The arrangement of the fusion splicer for end-view observation is shown; Figure 5 Two example double-nested anti-resonant nodeless optical fibers are shown in the side view images; Figure 6A and Figure 6B The diagram shows a sine wave obtained from an anti-resonant hollow fiber using two different cameras on a fusion splicer. Figure 7 A first example method for azimuth alignment of anti-resonant hollow optical fibers used for fusion splicing is shown. Figure 8 It shows the method for execution Figure 7 A schematic diagram of the first example apparatus of the method; Figure 9 It shows the use of Figure 7 The method used to obtain the data graph; Figure 10 It shows the method for execution Figure 7 A schematic diagram of a second example apparatus for the method; Figure 11 It shows the use of Figure 7 Methods and Figure 10 Data graphs acquired by the device; Figure 12 A second example method for azimuth alignment of anti-resonant hollow optical fibers used for fusion splicing is shown; Figure 13 It shows the use of Figure 12 The method used to obtain the data graph; Figure 14 A third example method for azimuth alignment of anti-resonant hollow optical fibers used for fusion splicing is shown. Figure 15 It shows the use of Figure 14 The method used to obtain the data graph; Figure 16 A fourth example method for azimuth alignment of anti-resonant hollow optical fibers used for fusion splicing is shown; and Figure 17 This is a block diagram of a fusion splicer configured to perform any of the methods described herein. The same reference numerals are used in the accompanying drawings to designate the same parts. Detailed Implementation

[0009] The specific embodiments provided below with reference to the accompanying drawings are intended to describe this example and are not intended to represent the only form in which this example can be constructed or utilized. This description illustrates the functionality of the example and the sequence of operations used to construct and operate the example. However, the same or equivalent functionality and sequences can be accomplished through different examples.

[0010] Figures 1 to 3 Cross-sectional views of three different examples of antiresonant hollow fiber (ARF) are shown. Light is guided in these fibers through antiresonant optical effects. Each fiber 100, fiber 200, and fiber 300 includes a tubular outer cladding (or sheath) 102, a structured inner cladding including multiple tubular cladding capillaries 104, 204, and 304, and a hollow core 106. The glass thickness of the outer cladding 102 is typically much greater than the thickness of the cladding capillaries 104, 204, and 304. Figure 1 In the first example shown, the structured inner cladding comprises five capillaries 104 having identical cross-sectional dimensions and shapes, arranged in a single ring within the outer cladding 102 such that each cladding capillary 104 is substantially parallel to the longitudinal axis of the outer cladding 102. Each cladding capillary 104 contacts (e.g., adheres to) the inner surface of the outer cladding 102 at an azimuth position 108, such that the cladding capillaries 104 are uniformly spaced around the inner circumference of the outer cladding 102 and are also spaced apart from each other by gaps 110 (i.e., such that there is no contact between adjacent capillaries). In some designs of ARF, the cladding capillaries 104 may be placed in contact with each other (in other words, unlike...). Figure 1 While the cladding capillaries are spaced apart as in the middle, eliminating this contact gap can improve the optical performance of the fiber. The gap 110 eliminates the nodes created by the contact points between adjacent tubes, which often lead to unwanted resonances and thus high losses. Therefore, optical fibers with spaced cladding capillaries can be called "node-free anti-resonant hollow-core fibers".

[0011] Cladding capillaries 104 are arranged in a ring around the interior of the tubular outer cladding 102, forming a central space, cavity, or void within the optical fiber. Their longitudinal axes are parallel to the longitudinal axes of both the outer cladding 102 and the cladding capillaries 104, forming the hollow core 106 of the optical fiber. The hollow core 106 is defined by the inward-facing portion of the outer surface of the cladding capillaries 104. This is the core boundary, and the capillary wall material (e.g., glass or polymer) constituting this boundary provides the desired anti-resonant photoconductive effect or mechanism. The cladding capillaries 104 have a thickness t at the core boundary, which defines the wavelength at which anti-resonant light guidance occurs in the ARF (Optical Resonance Response).

[0012] exist Figure 2 In the second example shown, each primary cladding capillary 104 contains a smaller secondary capillary 204 nested within it, bonded to the inner surface of the primary cladding capillary 104. In this example, its azimuth position 108 is the same as the bonding point between the primary cladding capillary 104 and the outer cladding 102. These additional smaller capillaries 204 can reduce optical loss. This type of ARF design with secondary capillaries can be referred to as "nested anti-resonant nodeless fiber" (NANF) (TM).

[0013] Figure 3 The third example shown incorporates two smaller clad capillary tubes 204 and 304 nested within each clad capillary tube 104. (Compared to...) Figure 2 As in the example shown, each smaller capillary 204, capillary 304 is bonded to the inner surface of the adjacent larger capillary, with its azimuth position matching the bonding point between the primary cladding capillary 104 and the outer cladding 102. In this example, the smaller capillary 304 can be referred to as the secondary cladding capillary, and the smallest capillary 204 can be referred to as the tertiary cladding capillary. The tertiary cladding capillary 204 is bonded to the inner surface of the secondary cladding capillary 304, which is bonded to the inner surface of the primary cladding capillary 104. This ARF design with secondary and tertiary cladding capillaries can be referred to as “double-nested anti-resonant nodeless fiber” (DNANF). In yet another example (not shown), the cladding capillaries may have a different configuration. For example, there may be smaller capillaries within the tertiary capillary 204 to provide further nesting levels and / or there may be multiple secondary capillary capillaries within each primary capillary, each secondary capillary capillary being bonded to the inner surface of the primary capillary capillary at different azimuth positions, and / or each primary capillary capillary may have internal structures (e.g., one or more partition walls).

[0014] Figures 1 to 3All examples shown include five primary cladding capillaries 104, thus exhibiting fivefold rotational symmetry. Furthermore, all cladding capillaries have circular cross-sections. In other examples, the number of primary cladding capillaries surrounding the core may vary (e.g., four, six, seven, eight, nine, or ten) and / or the cross-sections of the cladding capillaries may not be circular. Additionally, although in Figures 1 to 3 In the example, all primary cladding capillaries 104 have the same size and shape, but in other examples, the primary cladding capillaries within the outer cladding 102 may not all have the same size and / or shape.

[0015] Regardless of the precise details of the structure, from Figures 1 to 3 As clearly seen in the above description, ARF lacks continuous circular symmetry. When two ARF segments are fused together, any misalignment between the cladding capillaries (collectively referred to as the microstructures of the optical fiber) will increase the optical propagation loss at the splice point. Alignment requires careful rotational adjustments around the longitudinal axis of the fiber to align the structural features, which can be difficult and time-consuming.

[0016] Most existing fiber optic fusion splicers (or splicing devices) are not designed for use with ARF (Automatic Fiber Reinforced Radio) splicing, and therefore cannot perform the necessary azimuth alignment for low-loss splicing when splicing ARF. The few fusion splicers that can perform azimuth alignment and are therefore better suited for splicing ARF typically involve using a camera to observe or inspect the fiber endface. Figure 4 An example arrangement is shown. For end-view observation, a reflector 402 is temporarily inserted between the two end faces 404 and 406 to be fused. A light source 408 couples light into the fiber 410 to illuminate the end face 404, and the image is projected onto it by a camera 412 through the reflector 402 and lens 414. The fiber 410 can then be rotated to a predetermined orientation, followed by the use of a second light source 418 (which couples light into a second fiber 416) and a second reflector (…). Figure 4 (Not shown in the image) or the reflector 402 can be rotated 90° and this process repeated for the second fiber 416. Such devices are slow, expensive, precise, and bulky (e.g., due to the presence of a movable reflector), and are therefore primarily suited for indoor laboratory and cleanroom environments, rather than for field use. However, fiber optic splicing often needs to be performed in the field, such as during the installation of fiber optic communication networks. These applications require equipment that is as robust, portable, easy to operate, and fast as possible (e.g., because any optical cable may contain 60 or more fibers).

[0017] This paper describes a method and apparatus for achieving ARF rotational alignment solely through lateral fiber observation. By using lateral observation (rather than observing the end face), the need to insert into a reflector to observe the end face is avoided, resulting in a simpler, smaller, lower-cost, and more robust apparatus (e.g., due to fewer moving parts) and faster operation. When performing alignment in preparation for splicing two fibers, lateral observation allows the same camera to simultaneously observe both fibers using the same camera setup (something impossible with end-view observation), further simplifying the apparatus, reducing its cost, and enabling faster operation. The method described below can be used for any type of ARF (e.g., any ARF with greater than two-fold rotational symmetry), particularly for NANFs (e.g., such as...). Figure 2 (as shown) and DNANF (e.g., as shown) Figure 3 (As shown). The alignment method described herein can be used to perform rotational alignment of two ARF segments prior to fusion (e.g., fusion welding), or for other applications requiring precise rotational alignment of ARFs.

[0018] The method and apparatus described herein measure the intensity of pixel rows along the cross-section of an optical fiber in a captured ARF side-view image as a function of azimuth angle. This set of intensity data for multiple azimuth angles can be referred to as a sine curve. Reflection and refraction of the light beam as it passes through the ARF microstructure produce an intensity pattern in the camera that depends on the azimuth direction of the optical fiber. This pattern, which represents intensity variations along the cross-section of the optical fiber, is used to detect the azimuth direction of the fiber.

[0019] The terms “azimuth direction” and “azimuth rotation” are used in this document to refer to the angular (or rotational) direction and rotation of the ARF about its major axis.

[0020] Figure 5 The image shows two example DNANFs at random azimuth angles, image 502 (i.e., the left and right fibers have not yet been azimuth aligned). Figure 5 Also shown is a relative pixel intensity map 504 of the pixel rows along the fiber cross section (marked by white dashed line 506) in the captured side view image. Intensity map 504 shows peaks 508-514 at the outer and inner edges of the cladding 102, with a more complex intensity pattern 516 generated by the microstructure between these peaks.

[0021] Figure 6A The image shows a sine wave obtained from an ARF camera using two different cameras on a fusion splicer. The two different cameras are placed orthogonally to each other, thus capturing [the desired sine wave]. Figure 5Different side views of the fiber on the right (labeled X and Y). In these sine plots, the horizontal axis shows the pixel position in the pixel row along the fiber (pixel positions are labeled 0-N, where pixel 0 is the first pixel in the row and pixel N is the last pixel), the vertical axis shows the relative azimuth (the term "relative" is used because there is no defined angular reference position in the fiber), and the shades of gray represent the relative pixel intensity, so each horizontal line in the sine plot corresponds to the intensity at a different azimuth (plot 504).

[0022] Figure 6A The sine curve shown includes the raw pixel intensity data. Figure 6B The corresponding sine wave generated using the processed pixel intensity data is shown. The processed pixel intensity data is generated by processing the original pixel data; in the example shown, the processing includes low-pass filtering. Figure 6B In the example shown, the intensity distribution is essentially a sine curve. Other processing techniques can also be used instead of low-pass filtering (or other than low-pass filtering) to generate processed pixel intensity data. Examples of other processing techniques that can be used include averaging, spatial filtering, segmentation, Fourier transform, etc.

[0023] Although Figure 6A and Figure 6B The sine curve shown is for a full rotation of the fiber, but using the method described herein, the left and right fibers can be azimuthally aligned for splicing without requiring a full rotation of either or both fibers. This reduces the time required for fiber alignment, thereby shortening the splicing operation time. It may also reduce the mechanical complexity of the fusion splicer, as it does not require the ability to rotate either or both fibers a full 360°.

[0024] The method described herein involves aligning two ARFs simultaneously or separately. After alignment, the two ARFs are clamped and fixed, and can then be fused together, for example, using fusion splicing or mechanical splicing. The alignment process for each fiber involves side-illuminating the ARF and capturing intensity data for each pixel along a pixel row in the corresponding image of the side-illuminating ARF. The intensity data is processed and then analyzed to determine the pixel position in that row corresponding to a predefined feature (e.g., a maximum value, a minimum value, or a series of peaks and / or valleys in the processed intensity data). This determined pixel position corresponds to an initial azimuth direction, and depending on the specific method used, this process can be repeated to determine the pixel position of that feature in different azimuth directions. To align the two ARFs, one or both ARFs are rotated based on the determined pixel position of the predefined feature (e.g., in the initial azimuth direction, and in some examples, in other azimuth directions). Various different methods for using the determined pixel positions to determine the angle by which one or both ARFs should be rotated are described below.

[0025] Figure 7 A first example method for ARF azimuth alignment for fusion splicing is shown. This method is shown for a single fiber, but can be repeated for a second fiber (as indicated by the dashed arrows from block 714 to block 702) or performed substantially in parallel with the second fiber. This method can be used in situations such as... Figure 8 This is implemented on the fusion splicer shown.

[0026] like Figure 7 As shown, the ARF 802 with the cut end face 804 is inserted into the fusion splicer (block 702). The optical fiber can be placed in a removable fiber optic clamp before insertion into the fusion splicer (block 702), or alternatively, the optical fiber can be placed in the fiber optic clamp of the fusion splicer itself. The optical fiber (block 704) is then illuminated from the side using a light source 806. Intensity data for each pixel along a pixel row in the side view image captured by the camera 808 is captured and processed (block 705). This row of pixels passes through the optical fiber and is substantially perpendicular to the longitudinal axis of the fiber (e.g., as shown). Figure 5 (As shown by dashed line 506 in the diagram). The location of a feature or characteristic within that row of pixels in the processed intensity data (e.g., peak, valley, combination of peaks and / or valleys, or other features) is determined (block 706). The features used to determine the location (block 706) do not include any features from the cladding layer (e.g., features excluding those corresponding to the processed data). Figure 5 The original intensity data in Figure 504 shows any peaks 508-514. This pixel location (as determined by block 706) corresponds to the initial azimuth direction of the fiber. Figure 6B The initial azimuth direction α0 and the initial pixel position P0 of the feature are shown.

[0027] For this first iteration, after a single measurement in the initial azimuth direction, the fiber is then rotated to the new azimuth direction using a rotatable element 810 in the fusion splicer (block 708). The rotatable element 810 can rotate both the fiber and the fiber clamp simultaneously (with the fiber maintaining a fixed orientation relative to the fiber clamp), or the rotatable element 810 can rotate the fiber within the fiber clamp. Any suitable mechanical arrangement can be used as the rotatable element 810, capable of controlling the azimuth rotation of the fiber about its longitudinal axis (e.g., allowing it to rotate from a first angle to a second angle and then back to the first angle) while minimizing (or preventing) any lateral or longitudinal movement (i.e., movement along or perpendicular to the fiber's longitudinal axis).

[0028] Then, the method steps (blocks 704-706) are repeated in the new azimuth direction to determine the updated position of the same feature or property (i.e., the same feature used each time in block 706). The position determined in the second iteration corresponds to the new azimuth direction of the fiber (after rotation in block 708). The process of rotating the fiber (block 708) and determining the new pixel position with maximum intensity (block 710) is repeated until the peak pixel position of a specific feature or property is identified (block 712). Once the peak position is identified ("Yes" in block 712), the fiber is set to the azimuth corresponding to the identified peak position (this may require further rotation of the fiber in block 713, e.g., returning to the previous azimuth position) and clamped in place (block 714). Figure 6B The peak pixel location P of this feature is shown. P and the corresponding azimuth direction α P .

[0029] Determining the peak position (block 712) may include identifying the azimuth direction at which a specific feature or characteristic is located at the maximum pixel position. As mentioned above, the pixel numbers in the pixel row range from 0 to N, so the maximum pixel position is the pixel position with the highest number. Determining the peak position may involve additional processing of the pixel position data (from block 706), such as fitting a curve to the pixel position data (as described below). Figure 9 (as illustrated in the example).

[0030] In another example, the peak position can be defined relative to a predefined threshold, and the peak position (as determined in block 712) can be a position that exceeds (or is equal to or exceeds) that threshold. (See also:) Figure 6B In the example shown, the threshold can be set to pixel position 250. When using such a threshold, the number of iterations required before identifying the peak position may be reduced (and thus the alignment method may be faster) because there is no need to rotate beyond the peak position to identify that the peak position has been reached. This may also eliminate the step of repositioning the fiber (in block 713) after the peak position has been identified (in block 712).

[0031] Then, as Figure 7 The dashed arrow pointing from block 714 to block 702 indicates that the second ARF 812 can be repeated. Figure 7 The method, or the alignment of the second ARF 812, can be performed in parallel (e.g., substantially or partially parallel) with the alignment of the first ARF 802. The same features or characteristics used for the first ARF 802 are also used for the second ARF 812. Figure 8In the example shown, a separate light source 816, camera 818, and rotatable element 820 are provided for azimuth alignment of the second ARF 812. In other examples, a single light source and camera may be used for alignment of both the first ARF 802 and the second ARF 812. Once both ARF 802 and ARF 812 are aligned, the two ARFs can be fused together.

[0032] Although Figure 8 Not shown, but the fusion splicer may also include a control unit configured to perform Figure 7 The method comprises the following steps. A control unit is connected to light sources 806, 816, 808, 818, and rotatable elements 810 and 820, and is configured to control when each light source 806 and 816 illuminates, when each camera 808 and 818 captures an image, and how the rotatable elements 810 and 820 move to rotate ARFs 802 and 812 (in blocks 710 and 714). The control unit also includes image processing software configured to capture and process intensity data (in block 705), determine the pixel positions of features or characteristics (in block 706), and identify peak positions (in block 712). An example control unit will be provided in the reference... Figure 17 This will be described in more detail later. In various examples, the control unit can select the feature or characteristic used in the method. In other examples, the feature or characteristic may be predefined and the same for all ARFs (e.g., the highest intensity peak), or it may be selected based on user input.

[0033] Figure 9 It shows the use of Figure 7 The data graph obtained by the method (i.e., in multiple iterations of block 706) shows the relative azimuth angle on the horizontal axis and the pixel position of the feature or characteristic on the vertical axis. Figure 9 The diagram shows two lines—the characteristic location 902 in the processed intensity data (as determined in multiple iterations of block 706) and the fitted sine curve 904. Figure 9 In the example shown, at the initial azimuth angle α0, the feature (e.g., maximum intensity) is located at pixel position 175 (as determined in the first iteration block 706). The fiber is then rotated (in block 708), and the initial azimuth angle α0 and the final measured azimuth angle α... n Measurements were obtained at various relative angles between them (in subsequent iterations of block 706). The iterations of the method (blocks 704-708) did not exceed the final measured azimuth angle α. nThis is because it is clear from the processed data 902 that the peak pixel position has been identified ("Yes" in block 712). For example, as the fiber rotates, the feature position has increased from the initial pixel position 175 to nearly 180, and now has decreased to a pixel position below the initial pixel position 175. In this example, taking into account the noise in the processed intensity data 902, a fitted sine curve 904 is fitted to the processed intensity data 902 to identify the peak position used for the final positioning of the fiber (in blocks 713-714). Analysis of the fitted sine curve 904 identifies the relative angle α at which the peak pixel position appears. p Therefore, the fiber is moved back to that relative angle (in block 713) and clamped in place (in block 714). In other examples, the processed intensity data 902 can be used directly without curve fitting, or the processed intensity data 902 can be further processed in other ways.

[0034] like Figure 9 As shown, using Figure 7 When aligning using this method, it is not necessary to rotate the fiber a full 360°. For an ARF with n-fold rotational symmetry (where n>2), the maximum required rotation angle is slightly greater than 360° / n, as a peak is guaranteed within this angular range. By reducing the required angular displacement, the number of measurements can be reduced, thereby reducing the time required to perform alignment. Furthermore, if the amount of rotation required by rotatable elements 810 and 820 is less than 360°, the mechanical complexity of the fusion splicer can be reduced. Even if the fusion splicer is designed to accommodate any ARF with n-fold rotational symmetry (n>2), rotatable elements 810 and 820 only need to be able to rotate the fiber by a maximum of slightly more than 120° (e.g., ±60° from the center position, plus some additional small amplitudes), which reduces the mechanical complexity of rotatable elements 810 and 820.

[0035] Although the above description and Figure 7 The document addresses identifying the peak location of a feature (in block 712), but in other examples, it may also identify a valley value (i.e., the minimum location of the feature), or either a peak or valley value (in block 712). If a valley value (i.e., the minimum location) is used when aligning the first ARF 802, then a valley value must also be used when setting the azimuth alignment of the second ARF 812 (i.e., the other in the ARF pair to be fused together). The term "extreme pixel location" as used herein refers to a peak or valley value (i.e., the maximum or minimum location). When using a valley location (in block 712), a threshold can be used to identify this location (similar to when using a peak location); however, a valley location is considered any location where the value is less than a predefined threshold.

[0036] if Figure 7 The method identifies the peak or valley position when aligning the first fiber in the fiber pair to be spliced ​​(in block 712), thus reducing the amount of rotation required for the first fiber and the number of measurements performed; however, the rotation and number of measurements for the second fiber are unaffected. Therefore, this reduces the time required to perform alignment, but may not change the mechanical arrangement of rotatable elements 810 and 820 (e.g., because this requires that identical fibers with restricted rotational movement always be oriented first). Furthermore, when identifying the peak or valley for the first fiber, the final positioning of the second fiber (in blocks 713-714) cannot be performed entirely independently of the alignment of the first fiber (e.g., because the alignment of the second fiber requires knowing whether a peak or valley is to be identified in block 712).

[0037] exist Figure 8 In the arrangement shown, each fiber has a light source-camera pair (the pair includes a light source and a camera, arranged on opposite sides of the fiber); however, in other arrangements, there may be a single light source-camera pair for two fibers, or two or more light source-camera pairs for each fiber. Figure 10 An arrangement 1000 shown in cross-section is illustrated, comprising two light source-camera pairs (the first light source-camera pair includes light source 1002 and camera 1006, and the second light source-camera pair includes light source 1004 and camera 1008). The camera-light source pairs (1002 & 1006, 1004 & 1008) are arranged perpendicular to each other and may be referred to as capturing the X view (camera 1006) and the Y view (camera 1008). It is understood that there may be more than two light source-camera pairs in different directions, and even with multiple light source-camera pairs, it is not necessarily necessary to use all light source-camera pairs to perform the azimuth alignment of the method described herein (e.g., in some examples, only one light source-camera pair or a subset of light source-camera pairs may be used).

[0038] Figure 11 It shows the use of Figure 7 Methods and such Figure 10 The image shows data acquired by the two camera-light source pairs. Figure 9 The same as the diagram in the image, the horizontal axis shows the relative azimuth angle, and the vertical axis shows the pixel position of the feature or characteristic. Figure 11 Four lines are shown—the first line 1102 shows the feature locations in the processed intensity data from the X-view camera 1006 (as determined in multiple iterations of block 706), the second line 1106 shows the feature locations in the processed intensity data from the Y-view camera 1008, and the third and fourth lines 1104 and 1108 are sine curves fitted to the first and second lines, respectively. Although Figure 11The figure shows the complete 360° data, thus illustrating the five peaks caused by the fivefold symmetry of the ARF, but in the execution Figure 7 When using this method, it is not necessary to explore the entire angular range.

[0039] If two light source-camera pairs are used, the maximum angular rotation required for the first fiber to identify a peak in the feature location map of one of the cameras is reduced; however, the number of rotations and measurements for the second fiber remains unaffected. Therefore, this reduces the time required to perform alignment, but may not change the mechanical arrangement of rotatable elements 810 and 820 (e.g., because this requires that the same fiber with restricted rotational motion is always oriented first). Furthermore, whichever light source-camera pair (X or Y) is used to orient the first fiber (in block 712) must also be used for the second fiber (i.e., they must both use the X-view light source-camera pair, or both use the Y-view light source-camera pair). Moreover, when using either the X-view or Y-view light source-camera pair to identify the peak of the first fiber, the final positioning of the second fiber (in blocks 713-714) cannot be performed entirely independently of the alignment of the first fiber (e.g., because the alignment of the second fiber requires knowledge of which light source-camera pair is used). As mentioned above, by finding the peak or valley, the time required to align the first fiber can be further reduced because this halves the maximum amount of rotation required to locate the peak (or valley) of the first fiber, but it also limits the ability to align the second fiber completely independently of aligning the first fiber.

[0040] When multiple light source-camera pairs are present, regardless of whether one or more light source-camera pairs are used for azimuth alignment, the light sources can be selectively switched on and off so that only the light source in the light source-camera pair used to capture intensity data is turned on when capturing pixel intensity data (in block 705). By turning off all other light sources during measurement, the confusion effects caused by refraction or reflection from additional light sources in the internal microstructure of the ARF can be reduced.

[0041] Figure 12 A second example method for ARF azimuth alignment for fusion splicing is shown. This method involves using... Figure 7 The method involves initial coarse alignment of the two optical fibers (block 1202) followed by a subsequent fine alignment process (blocks 1204-1214). While coarse alignment (block 1202) can be performed sequentially on the first fiber and then on the second fiber, or simultaneously (e.g., depending on the light source-camera pair used and whether the peak / valley values ​​being sought are fixed), fine alignment involves directly comparing the side-view intensity maps of the two optical fibers.

[0042] like Figure 12 As shown, in use Figure 7 After coarse alignment of each fiber (block 1202), both fibers are illuminated from the side (block 1204). In the presence of multiple camera-light pairs, during fine alignment, a camera-light pair with the same orientation is used for each fiber (e.g., X-view or Y-view). The relative pixel intensity of pixel rows along the fiber cross-section in the captured image for each fiber is calculated (block 1206), for example, capturing images similar to... Figure 5 The diagram is shown. The fibers are laterally aligned using peaks 508-514 from the cladding (block 1208). During this lateral alignment, a camera-light source pair with X and Y views can be used for fine lateral alignment in both directions. One or both fibers are then rotated to improve the alignment between the respective intensity patterns 516 generated by the microstructure (block 1210). This can be referenced... Figure 13 The example shown is described below. The angular movement in the fine alignment operation (block 1210) is much smaller than that in the coarse alignment operation ( Figure 7 The angular displacement in ).

[0043] exist Figure 13 In the first figure 1300, the intensity patterns 1302 and 1304 from the two optical fibers do not completely overlap; however, by rotating and / or laterally moving the fibers (in the XY plane), they can overlap, as shown in the second figure 1306. Although Figure 13 The intensity maps are shown to be completely overlapping, but in other examples, fine rotational adjustments (block 1210) can reduce the difference between the two maps, although they may not completely overlap (e.g., due to manufacturing defects in one or the other fiber), and any suitable metric (e.g., alignment peak, mean square error, cross-correlation, FFT, etc.) can be used. After fine alignment is completed, the two fibers are clamped in place (block 1214) and ready for fusion splicing.

[0044] Figure 14 A third example method for ARF azimuth alignment for fusion splicing is illustrated. This method involves an initial calibration process 1400 for the fiber type (e.g., for fibers with specific internal microstructures), followed by an alignment process 1401, where calibration data is used to reduce the time required to perform alignment before fusion splicing compared to the methods described above. The calibration process 1400 only needs to be performed periodically and may only be performed once for each fiber type to generate calibration data.

[0045] The calibration process 1400 involves generating something similar to Figure 11 At least a portion of the figure shown. Although Figure 11 The image shows the characteristic position data of the two light source-camera pair and a complete 360° rotation, but... Figure 14The calibration process 1400 generates feature position data for one or more light source-camera pairs and at least 360° / (m x n) rotation, where m is the number of light source-camera pairs. For example... Figure 14 As shown, the method includes inserting an optical fiber for calibration into a fusion splicer (block 1402), and for each light source-camera pair and at a range of angular positions (span at least 360° / (m x n)), determining characteristic locations along the fiber cross-section in the processed intensity data (block 1406). The processed intensity data can be captured and processed as described above, and the characteristic locations can be identified as described above. This data is then stored (block 1408).

[0046] In use, after calibration, the first optical fiber to be fused is inserted into the fusion splicer (block 1410), and the first optical fiber is illuminated from the side using one or more light source-camera pairs arranged as during calibration. For each light source-camera pair, the characteristic position along the cross-section of the first optical fiber is determined by capturing, processing, and analyzing the intensity data of the pixel rows (block 1412). Based on this characteristic position data, a lookup is performed using calibration data to determine the current relative azimuth direction of the first optical fiber (block 1414). This method is also performed on the second optical fiber (blocks 1410-1414) to determine the current relative azimuth direction of the second optical fiber, which can be performed sequentially (as indicated by the arrow pointing from block 1414 to block 1410) or in parallel. After the relative azimuth directions of the two optical fibers to be aligned are determined, one or both fibers are rotated to align them in the same relative azimuth direction (block 1416), and the fibers are clamped and fixed (block 1418) in preparation for fusion splicing.

[0047] Figure 14 The azimuth direction is determined as a relative direction, not an absolute direction, because ARF exhibits rotational symmetry and lacks a predefined null position (unlike marked optical fibers). Furthermore, due to rotational symmetry, multiple azimuth directions can exist that give the same characteristic location. This can be referenced... Figure 15 Describe it. Figure 15 It shows Figure 11 The diagram corresponds to calibration data (as stored in block 1408). As mentioned above, the ARF has more than two-fold rotational symmetry; in the example shown, the ARF has five-fold rotational symmetry (n=5). This means that five different azimuth directions give the same feature location. Figure 15 In the image, the five positions with the feature location being pixel number 178 are marked by arrows 1501-1505.

[0048] Figure 14 For examples of how to use the method, please refer to [link / reference]. Figure 15The description is as follows. For example, if the feature position data (from block 1412) identifies a feature position of 175 from the X camera and a feature position of 165 from the Y camera for the first optical fiber, then the initial azimuth position α of the first optical fiber can be identified. 0,1 Similarly, if the feature location data (from block 1412) identifies the X-camera feature location of the second optical fiber as 171 and the Y-camera feature location as 147, then the initial azimuth position α of the second optical fiber can be determined. 0,2 Based on these two positions, the amount of rotation to be applied to either or both fibers can be determined (block 1414). There are several different methods to achieve this.

[0049] In the first example, it can be determined that the first optical fiber will be positioned from its initial azimuth position α. 0,1 Move to peak position α p The required rotation amount Δα1 (e.g., using Δα1=αp-α) 0,1 Similarly, it can be determined that the second fiber will be moved from its initial azimuth position α0,2 to its peak position α. p The required rotation amount Δα2 (e.g., using Δα2=α) p -α 0,2 Then each fiber is rotated by the determined angle Δα1 or Δα2 (block 1416) and clamped (block 1418) in preparation for fusion splicing. In the second example, once the initial azimuth positions of the two fibers are determined (block 1414), one of the fibers can be rotated by their initial angle difference Δα. 1,2 (For example, using Δα) 1,2 =α 0,2 -α 0,1 ) or Δα 2,1 The second angle difference is smaller because it takes into account the rotational symmetry of the ARF (e.g., considering the initial azimuth position α of the first fiber). 0,1 and α' 0,1 Given the same feature locations, they are equivalent.

[0050] Figure 16 A fourth example method for ARF azimuth alignment used in fusion splicing is shown. Figure 12 A variation of the method shown. In Figure 16 In the method, use Figure 14 The alignment process 1401 performs an initial coarse alignment (block 1602), and then, as described above, refers to... Figure 12 The fine alignment is performed.

[0051] While the methods described above involve capturing and processing intensity data for each pixel in a row of pixels in a side-view image captured by a camera (e.g., in block 705), in variations of any of the methods described above, intensity data for multiple rows of pixels (e.g., 10-20 rows of pixels) can be captured. When capturing intensity data for multiple rows of pixels, the raw intensity values ​​of the corresponding pixels in each row are averaged (e.g., for each pixel location i, where i = 0-N, the pixel intensity values ​​of that pixel are averaged across different rows) to create average intensity data for each pixel. The average intensity data is then processed (e.g., low-pass filtered) and used as described above. By averaging the pixel intensity values ​​of multiple rows of pixels, the methods described herein reduce sensitivity to defects and contamination (e.g., dirt on optical fibers, or small patches of coating that have not been completely stripped).

[0052] In all the methods described above, identifying the characteristic locations along the fiber cross-section (e.g., in any block 706, block 1206, block 1406, block 1412) excludes any features (e.g., peaks) in the processed intensity data corresponding to the cladding. Figure 5 Instead of using features of peak values ​​508-514 in the original intensity data shown in Figure 504, features generated by microstructures (e.g., features corresponding to peak values ​​508-514 in the processed intensity data) are used. Figure 5 (Characteristics of peak value 518 in the original intensity data).

[0053] Figure 17 This is a block diagram of the fusion splicer 1700 that performs the methods described herein. Figure 17The fusion splicer 1700 shown is a fusion fusion splicer; however, as described above, the methods described herein can be used for any type of fusion splicing. The fusion splicer 1700 includes two fiber clamps 1704 and 1714, which may be detachable or fixed within the fusion splicer 1700. The fusion splicer 1700 also includes two cameras 808 and 818, two light sources 806 and 816, two rotatable elements 810 and 820, and a fusion unit 1706 for performing the fusion. For mechanical fusion splicers, the fusion unit 1706 is omitted. The cameras 808 and 818 and the light sources 806 and 816 are arranged in light-camera pairs, with each camera in each pair positioned to capture light emitted from the light source in that pair after passing through the anti-resonant hollow fiber (ARF). As described above, the fusion splicer 1700 may have a single light-camera pair, or may include additional light-camera pairs (e.g., such that each fiber has two light-camera pairs). The fusion splicer also includes a control unit 1702 configured to perform the various steps of the method described above. The control unit 1702 is connected to light sources 806, 816, 808, 818, and rotatable elements 810 and 820, and is configured to control when each light source 806 and 816 is illuminated and its intensity, when each camera 808 and 818 captures an image and its gain, and how the rotatable elements 810 and 820 move to rotate ARF 802 and ARF 812 (e.g., in blocks 710 and 713).

[0054] The control unit 1702 includes one or more processors 1722, an input / output interface 1724 (through which the control unit communicates with other components in the fusion splicer), and a memory 1726. The memory 1726 stores software (processor-executable instructions) executed by the processors to implement the methods described above (e.g., image processing software 1728, fiber optic positioning software 1736, and fusion splice control software 1734). The memory 1726 also stores captured images (in an image repository 1730) and calibration map data 1732 (if used).

[0055] Image processing software 1728 is configured to: capture pixel intensity data of one or more rows of pixels in an image captured by camera 808 and camera 818; process the pixel intensity data (e.g., perform low-pass filtering on the raw intensity data); and determine the pixel positions of features based on the processed intensity data (in block 706) and identify peak positions (in block 712). Image processing software 1728 can also be configured to analyze differences between relative pixel intensity maps (in block 1210) and / or generate and / or use calibration data (in... Figure 14(In the method). Fiber optic positioning software 1736 is configured to control rotatable elements 810 and 820 to align the fiber (e.g., based on control signals received from image processing software 1728), and fusion splicing control software 1734 is configured to control fusion unit 1706 to perform fusion splicing once the fiber is aligned. Image processing software 1728 can also be further configured to analyze fusion quality after fusion is completed, such as fusion loss, microstructure misalignment, fiber tilt angle, and whether the fusion quality is acceptable or unacceptable.

[0056] In addition to the other examples described herein, or as alternatives, examples include any combination of the following terms:

[0057] Clause A: A method for azimuth alignment of two anti-resonant hollow fibers, comprising: for each anti-resonant hollow fiber: inserting the anti-resonant hollow fiber into a fusion splice (702, 1410); laterally illuminating the anti-resonant hollow fiber (704); capturing intensity data for each pixel in a pixel row of a corresponding image of the laterally illuminated anti-resonant hollow fiber, and processing the intensity data to generate processed intensity data (705); determining, based on the processed intensity data, a pixel position in the row corresponding to a predefined feature of the processed intensity data, the pixel position corresponding to an initial azimuth direction (706, 1412); and rotating at least one of the two anti-resonant hollow fibers based on the determined pixel position of the predefined feature (708-712, 1414-1416).

[0058] Clause B: The method described in Clause A, wherein processing the intensity data includes low-pass filtering the intensity data.

[0059] Clause C: The method described in Clause A or B, wherein the predefined features include peak values ​​in the processed intensity data.

[0060] Clause D: The method according to any of the preceding clauses, wherein rotating at least one of the two anti-resonant hollow fibers based on the determined pixel position of a predefined feature comprises, for each anti-resonant hollow fiber: rotating the anti-resonant hollow fiber to at least one new azimuth direction (708), and laterally illuminating the anti-resonant hollow fiber (704) at each new azimuth direction; capturing intensity data for each pixel in a pixel row of a corresponding image of the laterally illuminated anti-resonant hollow fiber at the new azimuth direction, and processing the intensity data to generate processed intensity data (705); determining, based on the processed intensity data, a pixel position in the row corresponding to a predefined feature, the pixel position corresponding to the new azimuth direction (705-706); identifying an extreme pixel position from the determined pixel positions (712); and clamping the anti-resonant hollow fiber in the azimuth direction corresponding to the extreme pixel position (714).

[0061] Clause E: The method according to Clause D, which captures intensity data for each pixel in a pixel row of a corresponding image of a side-illuminated antiresonant hollow fiber and processes the intensity data to generate processed intensity data, includes: for a first image and a second image of the side-illuminated antiresonant hollow fiber, capturing intensity data for each pixel in a pixel row of the image and processing the intensity data to generate processed intensity data, wherein the first image is captured using a first light source-image pair and the second image is captured using a second light source-image pair, and wherein determining the pixel position in the row corresponding to a predefined feature of the processed intensity data based on the processed intensity data includes: for the first image and the second image of the side-illuminated antiresonant hollow fiber, determining the pixel position in the row corresponding to a predefined feature of the processed intensity data based on the processed intensity data; and wherein identifying extreme pixel positions based on the determined pixel positions includes: identifying extreme pixel positions from pixel positions determined from images captured by a first camera or a second camera.

[0062] Clause F: The method described pursuant to Clause E further includes:

[0063] When using the first light source-camera pair to capture an image of a side-illuminated anti-resonant hollow fiber, the light source in the second light source-camera pair is turned off; and when using the second light source-camera pair to capture an image of a side-illuminated anti-resonant hollow fiber, the light source in the first light source-camera pair is turned off.

[0064] Clause G: The method according to any one of Clauses A to C, wherein rotating at least one of the two anti-resonant hollow fibers based on the determined pixel position of a predefined feature comprises: determining the current relative azimuth direction of the anti-resonant hollow fiber using a calibration map for each anti-resonant hollow fiber (1414), wherein the calibration map includes calibration data generated using another anti-resonant hollow fiber of the same type as the anti-resonant hollow fiber; rotating one or both anti-resonant hollow fibers to place them in the same relative azimuth direction (1416); and clamping the two anti-resonant hollow fibers (1418).

[0065] Clause H: The method according to Clause G further includes generating a calibration map by: inserting another anti-resonant hollow fiber into a fusion splicer (1402); laterally illuminating the other anti-resonant hollow fiber; for each of a plurality of azimuth directions of the other anti-resonant hollow fiber and for each of one or more light source-camera pairs: capturing intensity data for each pixel in a pixel row of an image of the laterally illuminated anti-resonant hollow fiber, and processing the intensity data to generate processed intensity data; and determining, based on the processed intensity data, the pixel position in the row corresponding to a predefined feature; and storing a map recording the pixel positions corresponding to the predefined features for each of the plurality of azimuth directions and for each light source-camera pair (1408).

[0066] Clause I: A method for azimuth alignment of two antiresonant hollow fibers, comprising: performing azimuth alignment on a first antiresonant hollow fiber and a second antiresonant hollow fiber using the method described in any of the preceding clauses; laterally illuminating the first and second antiresonant hollow fibers (1204); capturing an image of each laterally illuminated antiresonant hollow fiber and determining the relative pixel intensity of each antiresonant hollow fiber along a cross-section of the fiber (1206); laterally aligning the first and second antiresonant hollow fibers using the peak value in the relative pixel intensity caused by the cladding of the antiresonant hollow fiber (1208); rotating one or both antiresonant hollow fibers to improve the matching between the relative pixel intensities of portions corresponding to the microstructures of the antiresonant hollow fiber along the cross-section of the fiber (1210); and clamping the antiresonant hollow fiber (1214).

[0067] Clause J: A method for fusion splicing two anti-resonant hollow optical fibers together, the method comprising: azimuthally aligning the two anti-resonant hollow optical fibers using the method described in any of the preceding claims; and fusion splicing the aligned anti-resonant hollow optical fibers together.

[0068] Clause K: The method described in any of the preceding clauses, wherein the anti-resonant hollow fiber has more than twice the rotational symmetry.

[0069] Clause L: The method described under any of the foregoing clauses, wherein the antiresonant hollow fiber is a nested antiresonant nodeless fiber.

[0070] Clause M: The method described in any one of Clauses A through K, wherein the anti-resonant hollow fiber is a double-nested anti-resonant nodeless fiber.

[0071] Clause N: A fusion splicer (1700) for anti-resonant hollow fiber, comprising: a first rotatable element (810) for rotating a first anti-resonant hollow fiber through a plurality of azimuth directions; a second rotatable element (820) for rotating a second anti-resonant hollow fiber through a plurality of azimuth directions; a first light source (806, 1004) for side-illuminating the first anti-resonant hollow fiber; a first camera (808, 1008) for capturing an image of the side-illuminating first anti-resonant hollow fiber; and a control unit (1702) configured to: for each of the first and second anti-resonant hollow fibers: insert the anti-resonant hollow fiber into the fusion splicer. In the connector (702, 1410); the anti-resonant hollow fiber (704) is laterally illuminated; and intensity data is captured for each pixel in a pixel row of an image of the laterally illuminated anti-resonant hollow fiber, and the intensity data is processed to generate processed intensity data (705); and the pixel position in the row corresponding to a predefined feature of the processed intensity data is determined based on the processed intensity data, the pixel position corresponding to an initial azimuth direction (706, 1412); and at least one of the first and second anti-resonant hollow fibers is rotated based on the determined pixel position of the predefined feature (708-712, 1414-1416).

[0072] Clause O: The fusion splicer according to Clause N further includes a fusion unit (1702), wherein the control unit is further configured to: use the fusion unit to fusion the first anti-resonant hollow fiber and the second anti-resonant hollow fiber together.

[0073] Any range or device values ​​described in this document may be extended or changed without losing the desired effect, as will be apparent to those skilled in the art.

[0074] Although the subject matter has been described in language relating to structural features and / or method steps, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or steps described above. Rather, the specific features and steps described above are disclosed as examples of implementing the claims.

[0075] It should be understood that the benefits and advantages described above may relate to one embodiment or multiple embodiments. The embodiments are not limited to those that solve any or all of the described problems, nor are they limited to those that have any or all of the described benefits and advantages. It should also be understood that reference to "one" item means one or more of that item.

[0076] The operations described herein can be performed in any suitable order, or simultaneously when appropriate. Furthermore, individual steps can be removed from any method without departing from the scope of the subject matter described herein. Without sacrificing the desired effect, aspects of any of the above examples can be combined with aspects of the other examples to form further examples.

[0077] The term “comprising” as used herein means that the method steps or elements identified are included, but such steps or elements are not an exhaustive list, and a method or apparatus may include additional steps or elements.

[0078] The term “subset” as used in this article refers to a proper subset, which is a subset of a set that does not include all elements of the set (i.e., at least one element of the set is not included in the subset).

[0079] It should be understood that the above description is given by way of example only, and various modifications can be made by those skilled in the art. The above specification, examples, and data provide a complete description of the structure and use of exemplary embodiments. Although various embodiments have been described above with a degree of specificity or with reference to one or more individual embodiments, those skilled in the art can make numerous modifications to the disclosed embodiments without departing from the scope of this specification.

Claims

1. A method for azimuth alignment of two anti-resonant hollow optical fibers, comprising: For each antiresonant hollow fiber: Insert the anti-resonant hollow fiber into the fusion splicer; Lateral illumination is applied to the anti-resonant hollow optical fiber; Intensity data is captured for each pixel in a pixel row of a corresponding image of the antiresonant hollow fiber under side illumination, and the intensity data is processed to generate processed intensity data; and Based on the processed intensity data, determine the pixel position in the row corresponding to a predefined feature of the processed intensity data, wherein the pixel position corresponds to the initial azimuth direction; and Rotate at least one of the two anti-resonant hollow optical fibers based on the pixel position determined by the predefined features.

2. The method according to claim 1, wherein processing the intensity data comprises: The intensity data is then low-pass filtered.

3. The method according to claim 1 or 2, wherein the predefined feature includes: The peak value in the processed intensity data.

4. The method according to any one of the preceding claims, wherein rotating at least one of the two anti-resonant hollow fibers based on the pixel position determined by the predefined feature comprises, for each anti-resonant hollow fiber: The anti-resonant hollow fiber is rotated to at least one new azimuth direction, and at each new azimuth direction: Lateral illumination is applied to the anti-resonant hollow optical fiber; At the new azimuth direction, intensity data is captured for each pixel in a pixel row of the corresponding image of the anti-resonant hollow fiber under side illumination, and the intensity data is processed to generate processed intensity data; and The pixel position in the row corresponding to the predefined feature is determined based on the processed intensity data, and the pixel position corresponds to the new azimuth direction; The extreme pixel position is identified based on the determined pixel position; as well as The anti-resonant hollow fiber is clamped at the azimuth direction corresponding to the extreme pixel position.

5. The method of claim 4, wherein capturing intensity data for each pixel in a pixel row of a corresponding image of the antiresonant hollow fiber for side illumination and processing the intensity data to generate processed intensity data comprises: For each of the first and second images of the antiresonant hollow fiber under side illumination, intensity data is captured for each pixel in a pixel row of the image, and the intensity data is processed to generate processed intensity data, wherein the first image is captured using a first light source-image pair, and the second image is captured using a second light source-image pair. Furthermore, determining the pixel position in the row corresponding to the predefined feature of the processed intensity data based on the processed intensity data includes: For each of the first and second images of the anti-resonant hollow fiber under side illumination, the pixel position in the row corresponding to a predefined feature of the processed intensity data is determined based on the processed intensity data; Furthermore, identifying the extreme pixel position based on the determined pixel position includes: Extreme pixel positions are identified based on the pixel positions determined from the images captured by the first camera or the second camera.

6. The method of claim 5, further comprising: When the first light source-camera pair is used to capture an image of the side-illuminated anti-resonant hollow fiber, the light source in the second light source-camera pair is turned off; as well as When the second light source-camera pair is used to capture an image of the side-illuminated anti-resonant hollow optical fiber, the light source in the first light source-camera pair is turned off.

7. The method according to any one of claims 1 to 3, wherein rotating at least one of the two anti-resonant hollow optical fibers based on the pixel position determined by the predefined feature comprises: For each antiresonant hollow fiber, a calibration map is used to determine the current relative azimuth direction of the antiresonant hollow fiber, wherein the calibration map includes calibration data generated using another antiresonant hollow fiber of the same type as the antiresonant hollow fiber. Rotate one or both of the anti-resonant hollow fibers so that they are in the same relative azimuth direction; and Two anti-resonant hollow optical fibers are clamped together.

8. The method of claim 7, further comprising generating the calibration map by: Insert the other anti-resonant hollow fiber into the fusion splicer; Lateral illumination is applied to the other anti-resonant hollow fiber; For each of the multiple azimuth directions of the other anti-resonant hollow fiber and for each of the one or more light source-camera pairs: Intensity data is captured for each pixel in a pixel row of an image of the antiresonant hollow fiber under side illumination, and the intensity data is processed to generate processed intensity data; as well as The pixel position in the row corresponding to the predefined feature is determined based on the processed intensity data; as well as For each of the plurality of azimuth directions and for each light source-camera pair, a map is stored showing the pixel positions corresponding to the predefined features.

9. A method for azimuth alignment of two anti-resonant hollow optical fibers, comprising: The azimuth alignment of the first anti-resonant hollow fiber and the second anti-resonant hollow fiber is performed using the method of any one of the preceding claims; Lateral illumination is applied to both the first anti-resonant hollow fiber and the second anti-resonant hollow fiber. Capture images of each side-illuminated anti-resonant hollow fiber and determine the relative pixel intensity along the fiber cross-section for each anti-resonant hollow fiber; The first and second anti-resonant hollow fibers are laterally aligned using the peak value in the relative pixel intensity caused by the cladding of the anti-resonant hollow fiber. Rotate one or both of the anti-resonant hollow fibers to improve the matching between the relative pixel intensities of the portions of the fiber cross-section corresponding to the microstructure of the anti-resonant hollow fiber; as well as The anti-resonant hollow fiber is clamped.

10. A method for fusion splicing two anti-resonant hollow optical fibers together, the method comprising: The azimuth alignment of two anti-resonant hollow optical fibers is performed using the method described in any one of the preceding claims; as well as The aligned anti-resonant hollow optical fibers are fused together.

11. The method according to any one of the preceding claims, wherein the anti-resonant hollow fiber has a rotational symmetry greater than twofold.

12. The method according to any one of the preceding claims, wherein the anti-resonant hollow fiber is a nested anti-resonant nodeless fiber.

13. The method according to any one of claims 1 to 11, wherein the anti-resonant hollow fiber is a double-nested anti-resonant nodeless fiber.

14. A fusion splicer for anti-resonant hollow optical fiber, comprising: A first rotatable element is used to rotate a first anti-resonant hollow optical fiber through multiple azimuth directions; The second rotatable element is used to rotate the second anti-resonant hollow fiber through multiple azimuth directions; A first light source is used to provide lateral illumination to the first anti-resonant hollow optical fiber. A first camera is used to capture an image of the first anti-resonant hollow optical fiber under side illumination. as well as Control unit, the control unit being configured to: For each of the first and second anti-resonant hollow fibers: Insert the anti-resonant hollow fiber into the fusion splicer; Lateral illumination is applied to the anti-resonant hollow fiber; and Intensity data is captured for each pixel in a pixel row of an image of the antiresonant hollow fiber under side illumination, and the intensity data is processed to generate processed intensity data; as well as Based on the processed intensity data, determine the pixel position in the row corresponding to a predefined feature of the processed intensity data, wherein the pixel position corresponds to the initial azimuth direction; and Rotate at least one of the first and second anti-resonant hollow optical fibers based on the pixel position determined by the predefined features.

15. The fusion splicer of claim 14, further comprising a fusion unit, wherein the control unit is further configured to: The first anti-resonant hollow fiber and the second anti-resonant hollow fiber are fused together using the fusion unit.

Images