Optical device and coherent receiver

Abstract

To provide an optical device and a coherent receiver that can reduce the number of photo detection elements for obtaining two complex amplitudes of signal light regardless of the polarization state of local light.SOLUTION: A coherent receiver 10 includes a light wave converter 21, a light-receiving unit 14, and a processing unit 13. The light wave converter 21 is composed of two meta-surfaces 25 and 26 and converts input local light Lo into five split local lights, each of which is combined with signal light Sig and enters a light-receiving surface 14a of the light-receiving unit 14. The interference light between each split local light and the signal light Sig is received by a photo detection element 19 provided on the light-receiving surface 14a. The meta-surfaces 25 and 26 apply polarization conversion and phase change to each split local light to make it have different polarization states so that conditions for decoding the signal light are met regardless of the polarization state of the local light Lo.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an optical device and a coherent receiver. [Background technology]

[0002] A coherent receiver detects the two complex amplitudes of the signal light to obtain the information carried by the signal light. This coherent receiver detects the complex amplitude by causing interference between the signal light and the local light. Known coherent optical communications using such coherent receivers include the general coherent method, in which the receiving side (i.e., the coherent receiver) has a local light source, and the self-coherent method, in which the transmitting side sends the local light along with the signal light to the receiving side via an optical fiber. The self-coherent method has the advantage that the transmitting side can use the same light source (e.g., a laser device) for the signal light and the local light, eliminating the need for wavelength tuning of the signal light and the local light.

[0003] The polarization state of the local light must be fixed. For this reason, self-coherent systems include systems that are equipped with a circuit that tracks and compensates for fluctuations in the polarization state of the local light due to transmission through an optical fiber, and systems that maintain the polarization state of the local light while transmitting using a polarization-maintaining fiber (e.g., Non-Patent Documents 1 and 2). Self-coherent coherent receivers that can operate with local light in any polarization state are also known (Non-Patent Documents 3 and 4). The coherent receivers in Non-Patent Documents 3 and 4 use five or six balanced photodetectors. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] T. Gui et al., “Real Time 6.4 Tbps (8×800G) SHCD Transmission through 1+8 Multicore Fiber for Co-Packaged Optical-IO Switch Applications,” OFC Post-Deadline Paper, Th4C.1 (2022). [Non-patent document 2] T. Gui et al., “Real-Time Demonstration of 600 Gb / s DP-64QAM Self Homodyne Coherent Bi-Direction Transmission with Un-Cooled DFB Laser,” OFC Post-Deadline Paper, Th4C.3 (2020). [Non-patent document 3] H. Ji et al., “Polarization-diversity receiver using remotely delivered local oscillator without optical polarization control,” Opt. Express, 28(15), 22882-22890 (2020). [Non-patent document 4] H. Ji et al., “Photonic Integrated Self-Coherent Homodyne Receiver Without Optical Polarization Control for Polarization-Multiplexing Short-Reach Optical Interconnects,” J. Light. Technol., 41(3), 911-918 (2023). Summary of the Invention [Problem to be solved by the invention]

[0005] Incidentally, a balanced photodetector is composed of a pair of photodetector elements, and the coherent receivers described in Non-Patent Documents 3 and 4 above require 10 to 12 photodetector elements.

[0006] The present invention has been made in consideration of the above circumstances, and aims to provide an optical device and a coherent receiver that can reduce the number of photodetectors required to obtain two complex amplitudes of signal light regardless of the polarization state of the local light. [Means for solving the problem]

[0007] The optical device of the present invention is such that N is an integer equal to or greater than 5, and the Jones matrix of the signal light to the nth (n=1, 2, . . . , N) light receiving position is expressed as J n , the local light transformation Jones matrix K n The optical fiber includes a lightwave converter that converts at least one of the signal light and the local light into at least N converted lights with different polarization states by applying polarization conversion and phase change so as to satisfy conditional formula (I), and causes the signal light and the local light to interfere at N light-receiving positions with at least one of the lights as converted light. † " denotes the Hermitian transpose.

number

[0008] A coherent receiver according to the present invention comprises the above optical device and a processing unit that obtains the complex amplitude of the signal light based on the light reception results of the plurality of photodetector elements. [Effects of the Invention]

[0009] According to the present invention, N is an integer equal to or greater than 5, and the Jones matrix J of the signal light to the nth (n=1, 2, . . . , N) light receiving position is n and the local light transformation Jones matrix K n The matrix H determined from nAt least one of the signal light and the local light is subjected to polarization conversion and phase change to convert them into polarization states different from each other so that a predetermined conditional expression is satisfied, and the signal light and the local light are made to interfere with each other using at least one of the lights as converted light. Therefore, it is possible to obtain two complex amplitudes of the signal light regardless of the polarization state of the local light with fewer detection elements that receive the interference light resulting from this interference. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is an explanatory diagram illustrating a configuration of a coherent receiver according to a first embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of a circuit configuration in a processing unit. [Figure 3] This is a perspective view showing the appearance of a meta-atom provided on a metasurface. [Figure 4] An explanatory diagram showing the arrangement of meta-atoms in a metasurface. [Figure 5] FIG. 2 is an explanatory diagram illustrating the function of a light wave converter. [Figure 6] 1 is a graph showing the sensitivity of a coherent receiver to the polarization state of the local light. [Figure 7] 1 is a graph showing the sensitivity of a conventional local optical polarization dependent coherent receiver to the polarization state of the local light. [Figure 8] FIG. 10 is an explanatory diagram showing the configuration of a coherent receiver according to a second embodiment in which signal light and local light are input to a lightwave converter made up of a single metasurface. [Figure 9] FIG. 10 is an explanatory diagram illustrating the function of a lightwave converter in a coherent receiver according to the second embodiment. [Figure 10] FIG. 10 is an explanatory diagram showing the configuration of a coherent receiver according to a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] [First embodiment] 1, the coherent receiver 10 includes an optical device 11 and a processing unit 13. The optical device 11 generates a plurality of interference lights by causing input signal light Sig and local light Lo to interfere with each other under predetermined conditions, and receives each interference light at a light receiving unit 14. The coherent receiver 10 obtains the complex amplitude of the signal light Sig at the processing unit 13 based on the light receiving result at the light receiving unit 14.

[0012] The coherent receiver 10 receives the signal light Sig from the optical fiber 17 and the local light Lo from the optical fiber 18. The optical fibers 17 and 18 are single-mode fibers. The signal light Sig is quadrature modulated (IQ modulated) based on the data to be transmitted. The coherent receiver 10 is configured to be independent of the local light polarization and can detect the signal light Sig regardless of the polarization state (electric field oscillation direction and phase difference) of the input local light Lo. In this example, the coherent receiver 10 uses a homodyne detection method, and the wavelength of the local light Lo is the same as the wavelength of the signal light.

[0013] That is, in this coherent receiver 10, the polarization state of the local light Lo may change from a specific state or may fluctuate over time during transmission from the transmitting side via, for example, an optical fiber 18. Such a coherent receiver 10 is preferably used, for example, as part of a self-coherent coherent optical communication system in which the local light Lo is sent from the transmitting side to the receiving side via an optical fiber. It is also useful in cases other than the self-coherent system where a light source for the local light Lo is present on the receiving side and the polarization state of the local light Lo fluctuates. In other words, it is permissible to use a light source whose polarization state fluctuates as the light source for the local light Lo.

[0014] As will be described later, the optical device 11 is configured to receive (detect) interference light using 5 (=N) photodetecting elements (photodetecting elements) 19 of the light receiving section 14. "N" is the number of interference light beams generated by the optical device 11 for coherent reception, and also the number of photodetecting elements 19 that receive the interference light beams. In this example, N=5, but N can be an integer equal to or greater than 5, and is preferably within the range of 5 to 9.

[0015] In addition to the light receiving unit 14, the optical device 11 also includes a lightwave converter 21, a lens 22, and a multiplexer 23. The lightwave converter 21 is composed of two metasurfaces 25 and 26. Both metasurfaces 25 and 26 are transmissive and control the amplitude distribution, phase distribution, and polarization state distribution of the transmitted light. The metasurfaces 25 and 26 are, for example, circular plate-shaped with a diameter of about 1 mm, and the metasurfaces 25 and 26 are arranged on top of each other so that the incident surface of the metasurface 26 is in contact with the exit surface of the metasurface 25.

[0016] In this example, a lightwave converter 21, i.e., metasurfaces 25 and 26, is arranged on the optical path of the local light Lo emitted from the optical fiber 18. As the local light Lo passes through the metasurfaces 25 and 26 sequentially, the local light Lo is converted into five converted lights (hereinafter referred to as converted local lights) that have been given polarization conversion and phase changes. Therefore, in this example, the number of interference lights generated by the lightwave converter 21 is the same as the number of converted lights generated from a single input light. As will be described in detail later, the lightwave converter 21 gives polarization conversion and phase changes to each converted local light so that the polarization states of each converted local light are different from each other. The lightwave converter 21 converts (splits) the local light Lo so that the optical intensities of each converted local light are equal. Note that the converted local lights can have different optical intensities, but for simplicity of processing, it is preferable to make the optical intensities of each converted local light equal, as in this example.

[0017] Constructing the lightwave converter 21 using metasurfaces 25 and 26 as described above is advantageous for miniaturizing the optical device 11 and, ultimately, the coherent receiver 10. In this example, transmissive metasurfaces 25 and 26 are used as described above, but reflective metasurfaces may also be used, or a combination of reflective and transmissive metasurfaces may be used. Furthermore, the lightwave converter 21 may also be constructed using three or more metasurfaces.

[0018] For example, the local light Lo is incident on a predetermined incident region with a predetermined spread from the normal direction of the incident surface of the metasurface 25. Each converted local light emitted from the metasurface 26 passes through the multiplexer 23 and is incident on the light-receiving surface 14a of the light-receiving unit 14. The lightwave converter 21 focuses each converted local light at a different position on the light-receiving surface 14a. In this example, the positions where the converted local light is focused are arranged in a line, but the arrangement is not limited to this and may be, for example, a two-dimensional arrangement.

[0019] The signal light Sig is incident on the optical device 11 from a direction perpendicular to the incident direction of the local light Lo. A lens 22 and a multiplexer 23 are arranged on the optical path of the signal light Sig, and the signal light Sig that has passed through the lens 22 is reflected by the multiplexer 23 and enters the light-receiving surface 14a. The lens 22 is a positive lens and focuses the signal light Sig on the light-receiving unit 14.

[0020] The multiplexer 23 is composed of a half mirror and superimposes and combines each converted local light from the lightwave converter 21 that passes through it with the reflected signal light Sig. This multiplexer 23 only needs to be able to superimpose the split local light and the signal light Sig, and for example, a polarization-independent beam splitter may be used. The lens 22 may also be a metasurface that focuses light by controlling the in-plane phase distribution.

[0021] In this example, the local light Lo is incident on the lightwave converter 21 to impart polarization conversion and a phase change, but instead of the local light Lo, the signal light Sig may be incident on the lightwave converter 21, and the signal light Sig may be converted into converted light that has been subjected to polarization conversion and a phase change. In this case, the local light Lo is configured to be incident on the light-receiving unit 14 via the lens 22 and the multiplexer 23. Furthermore, as will be described later, it is also possible to configure the signal light Sig and the local light Lo to be subjected to polarization conversion and a phase change, respectively.

[0022] As described above, the light receiving unit 14 has five photodetecting elements 19, which are arranged in a line at predetermined intervals on the light receiving surface 14a. Each converted local light from the lightwave converter 21 is incident on each photodetecting element 19 so that one converted local light is condensed and incident on one photodetecting element 19. The beam diameter of the signal light Sig on the light receiving surface 14a is adjusted by the lens 22 so that the signal light Sig is incident on each photodetecting element 19. As a result, signal light Sig of the same optical intensity is incident on each photodetecting element 19.

[0023] As described above, by irradiating the signal light Sig and the converted local light on each photodetector element 19, they interfere with each other to generate interference light, which is then received by each photodetector element 19. That is, each photodetector element 19 is disposed at a light-receiving position where it receives the interference light. The photodetector element 19 is formed of, for example, a photodiode, receives the interference light, and converts the light intensity into an electrical light-receiving signal.

[0024] In the following description, when distinguishing between the photodetecting elements 19, they will be referred to as the first photodetecting element 19, the second photodetecting element 19, etc. in the order in which they are arranged on the light-receiving surface 14a for convenience. Similarly, when distinguishing between interference lights, they will be described in the order in which they are arranged (the order in which they are incident on the light-receiving surface 14a and received). When distinguishing between converted lights (in this example, converted local light), they will be described in the order in which they are arranged (the order in which they are incident on the light-receiving surface 14a for convenience. In this example, the order in which the incident positions of one converted local light are incident matches the order in which the light-receiving positions of the interference light generated from that converted local light are received. Furthermore, when distinguishing between received light signals, numbers (ordinal numbers) corresponding to the photodetecting elements 19 will be assigned.

[0025] The processing unit 13 obtains the complex amplitude of the X polarization and the complex amplitude of the Y polarization of the signal light Sig from the received light signals of the five interference lights. For example, as shown in FIG. 2, the processing unit 13 is composed of four differential amplifier circuits 28, a digital calculation unit 29, etc. Each differential amplifier circuit 28 obtains a differential signal between the first and second received light signals, a differential signal between the second and third received light signals, a differential signal between the third and fourth received light signals, and a differential signal between the fourth and fifth received light signals. The digital calculation unit 29 performs calculations using the digitally converted differential signals to obtain the complex amplitude (A XI , A XQ ) and the complex amplitude of the Y polarization (A YI , A YQ ) respectively.

[0026] The metasurfaces 25 and 26 described above are composed of a plurality of metaatoms (scatterers) of minute size (subwavelength order) arranged on a two-dimensional plane, which locally change the phase, amplitude, and polarization of incident light. The two-dimensional distribution of the phase, amplitude, and polarization of light can be controlled by adjusting the shape of each metaatom.

[0027] As shown in Figure 3, one metaatom 31 on the metasurface 25 is provided so as to protrude from the surface of a transparent substrate 32. In this example, the metaatom 31 has an elliptical cylindrical shape. As shown in Figure 4, the metaatoms 31 in the metasurface 25 are arranged in a square array on the substrate 32, with each metaatom 31 located at each vertex of a square. The metaatoms 31 are made of a dielectric material (e.g., silicon). While the metaatoms 31 may be made of other materials, a metasurface 25 with metaatoms 31 formed of a dielectric material has low optical loss and is advantageous for constructing a coherent receiver 10. The substrate 32 is made of a material with high transmittance for the local light Lo, such as a glass substrate or a quartz substrate. The period Λ of the metaatoms 31 (the length of one side of the square in the square array) is set to be equal to or less than the wavelength of the signal light Sig.

[0028] The shape of the meta-atom 31 is not limited to an elliptical cylinder. For example, it may be a prism with a rectangular or rhombic cross section. Also, in this example, the meta-atoms 31 are arranged in a square, but the arrangement is not limited to this and may be, for example, a regular triangular arrangement.

[0029] The shape of the meta atom 31 provided as above, that is, the length of the major axis Da, the length of the minor axis Db, and the inclination of the major axis θ MA , height h MA By adjusting the phase of the light incident on the position of the meta-atom 31, birefringence can be imparted to the light incident on the position of the meta-atom 31 with respect to any optical axis. In other words, it is possible to control the phase of the light polarized in the long axis direction and the phase of the light polarized in the short axis direction of the light incident on the position of the meta-atom 31. As a result, the metasurface 25 can impart polarization conversion and phase change to the light passing through it, and by controlling the in-plane phase distribution, it is possible to split and focus the light. Note that the metasurface 26 has a similar configuration to the metasurface 25, so a description thereof will be omitted.

[0030] To determine the shape of each metaatom on the metasurfaces 25 and 26, we first define the distribution of the Jones matrix corresponding to the transformation from the input of the local light Lo to the light-receiving surface 14a. Then, we calculate the distribution of the Jones matrix on the output surface of the metasurface 26 by performing backpropagation calculations on the four elements of the Jones matrix. The backpropagation calculations can be performed using angular spectrum analysis or other methods. The Jones matrix from the optical fiber 18 to the input surface of the metasurface 25 is calculated in the same way, and the Jones matrix at each point on the metasurfaces 25 and 26 is obtained from the ratio of these. The shape of each metaatom can be determined based on the Jones matrix obtained in this way. The method for determining the shape of each metaatom from the Jones matrix is ​​described in detail, for example, in the literature "B. Mirzapourbeinekalaye, et al., "General Lossless Polarization and Phase Transformation Using Bilayer Metasurfaces," Adv. Opt. Mater., 10, 2102591 (2022)."

[0031] The lightwave converter 21 applies polarization conversion and phase change to each converted local light to make it have different polarization states with respect to the nth (n=1, 2, N) interference light so as to satisfy conditional formula (I). This makes it possible to obtain the complex amplitudes of the X-polarized wave and the Y-polarized wave of the signal light Sig independently of the polarization state of the local light Lo. n is the Jones matrix of the signal light Sig to the nth receiving position, K n is the Jones matrix of the transformation of the local light Lo to the n-th light receiving position, and H n are the Jones matrices J n , K. n The Jones matrix is ​​expressed as follows: † " denotes the Hermitian transpose.

[0032]

number

[0033] As shown in equation (1), the electric field of the signal light Sig is expressed by a vector (Jones vector) A, the electric field of the local light Lo is expressed by a vector (Jones vector) B, and the angular frequencies of the signal light Sig and the local light Lo are expressed by ω C The electric field of the n-th interference light incident on the n-th photodetector element 19 is E n When this is done, the electric field E n is the Jones matrix transformation J n , K. n This is expressed as equation (2) using

[0034]

number

[0035] On the other hand, in order to maximize the receiving sensitivity of the coherent receiver 10, the transformation Jones matrix H k , H h (k≠h) is the vector in the first column of H k1 , H h1 When N=5, the value Rs given by equation (II) is set to the minimum value of "-1 / 4". As will be described later, in a configuration where N=5, it is known that the theoretical maximum receiving sensitivity can be achieved by setting the value Rs to the minimum value of "-1 / 4". Note that "< , >" in equation (II) is a Hermite inner product.

[0036]

number

[0037] As in this example, when the lightwave converter 21 is not applied to the signal light Sig but is applied only to the local light Lo, the transformation Jones matrix H n can be expressed as in equation (3). In this case, the transformation Jones matrix J n , the Jones matrix K for the local light Lo n is expressed as in equation (4). The transformation Jones matrix J for the signal light Sig in equation (4) is nmeans that the lightwave converter 21 does not act on the signal light Sig. On the other hand, the transformation Jones matrix K n is the light obtained by dividing the local light Lo that becomes the nth converted local light, and is divided by an angle θ from the X and Y axes. n Polarization conversion by adding a phase difference π between the polarization components in the rotated direction, and the shift amount φ to both the X and Y polarization components 0n and the phase difference Δφ between the X-polarized and Y-polarized components. n In addition, even when the lightwave converter 21 acts only on the signal light Sig and does not act on the local light Lo, the converted Jones matrix H n is expressed similarly.

[0038]

number

[0039] By applying the formula (3) to the above conditional formula (I), the condition for achieving polarization independence is expressed as the formula (5).

number

[0040] The conversion by the light wave converter 21 in this example is performed by the angle θ n , phase φ 0n , phase difference Δφ n The transformed Jones matrix K is given by n Each value in equation (6) is a solution in which the constant (const.) in equation (5) is set to the value "0" and the value Rs is set to "-1 / 4".

[0041]

number

[0042] The functions of the lightwave converter 21 configured as above are shown in Fig. 5. The lightwave converter 21 functions as a beam splitter 34, first retarders (retarders) 351 to 355, phase shifters 361 to 365, second retarders 371 to 375, and condenser lenses 381 to 385. The first retarder 35 n , phase shifter 36 n , second phase shifter 37 n and condenser lens 38 n corresponds to the beam splitter 34. The beam splitter 34 splits the incident local light Lo into five beams of light. These split beams are subjected to polarization conversion and phase change to generate converted local light. In the following description, the traveling direction of the local light Lo is defined as the Z axis, and two axes that are perpendicular to the Z axis and perpendicular to each other are defined as the X axis and the Y axis. The tilt of the optical axis in each function is defined as the tilt with respect to the X axis, with the X axis being the reference (0°). The direction of the X axis can be set arbitrarily as long as it is perpendicular to the Z axis.

[0043] 1st phase shifter 35 n is the angle θ n The phase shifter 36 corresponds to a half-wave plate with an inclined optical axis, and converts the polarization by advancing the phase of the polarization component perpendicular to the optical axis by a phase difference of π with respect to the polarization component in the optical axis direction. n is used to shift the phase of the split light, and the phases of the two orthogonal polarization components are each set to the same magnitude (phase φ 0n ) shifts the phase. n The magnitude of the phase shift due to the second retarder 37 is changed for each converted local light, thereby shifting the phases of the converted local lights. n The second retarder 37 converts the polarization by providing a phase difference with the tilt of the optical axis set to 0°. n is the phase difference Δφ between the Y polarization component and the X polarization component. n Focusing lens 38 n The converted local light is focused on the photodetector element 19. In this example, the first retarder 35 n and second retarder 37 n provides the polarization transformation, and the phase shifter 36 n gives the phase change.

[0044] In this example, the metasurface 25 has the functions of the beam splitter 34, first retarders 351 to 355, and phase shifters 361 to 365, and the metasurface 26 has the functions of the second retarders 371 to 375 and condenser lenses 381 to 385. The first retarders 351 to 355 and the second retarders 371 to 375 must impart phase differences in this order, but the phase shifts by the phase shifters 361 to 365 may be performed before the polarization conversion by the first retarders 351 to 355, or may be performed after the polarization conversion by the second retarders 371 to 375.

[0045] In order to realize the function of the above-mentioned light wave converter 21 with a simple cylindrical meta-atom 31 such as an elliptical cylindrical shape that has no twist in the direction of light propagation, the polarization conversion functions of the first retarders 351 to 355 and the second retarders 371 to 375 are divided into metasurfaces 25 and 26.

[0046] It is also possible to provide a metasurface for each function, or to realize one function using multiple metasurfaces. Of course, as in this example, using the minimum number of metasurfaces is preferable in order to reduce light loss and the number of parts.

[0047] The lightwave converter 21 can also be constructed using optical elements such as a beam splitter, a retarder, a phase shifter, and a focusing lens, or can be constructed by combining such optical elements with a metasurface. On the other hand, if the lightwave converter 21 is constructed using a metasurface, as in this example, the lightwave converter 21 can be realized with two metasurfaces 25 and 26, which allows for an extremely simple configuration of the lightwave converter 21 and is advantageous for miniaturization.

[0048] Although not limited to the configuration of this example, where N is an integer of 5 or more, a photocurrent I n can be expressed as in equation (7) from equation (2). Note that the value R P indicates the sensitivity of the photodetector element 19.

[0049]

number

[0050] The value I in the above formula (7) Sn , I Bn , I Ln is the photocurrent I n The photocurrent I corresponds to the beat component between the signal light and the local light (converted local light in this example), and the beat component between the local light and the local light. n where the photocurrent I n , current component I Sn , I Bn , I Ln The column vectors are the component vectors I and I S , I B , I L Among these component vectors, component vector I B is a linear component of the signal light Sig, and is a component required for demodulation, i.e., to obtain complex amplitude. On the other hand, the component vector I S is nonlinear with respect to the signal light Sig and needs to be removed. L can be ignored because it is a constant (DC component).

[0051] As with general balanced detection, the component vector I S When extracting the four signal components S by removing the 4 rows and N columns, the matrix that specifies how to extract the difference in the photocurrent is defined as "C", and the matrix C can be expressed as in equation (8). S = 0". Note that the component vector I S can be expressed as equation (8a).

[0052]

number

[0053] Here, the transformed Jones matrix J is used to apply polarization transformation and phase change to the signal light Sig. n is expressed as the product of a unitary matrix and a diagonal matrix as in equation (9). n This also applies to 90° optical hybrid circuits, 120° optical hybrid circuits, 72° optical hybrid circuits, etc., used in coherent receivers.

[0054]

number

[0055] Transformed Jones matrix J n When expressed as equation (9), the current component I between the signal light and the signal light is Sn can be expressed as in equation (10), and from this result, the matrix C must satisfy the requirements of equation (11). T " is the transpose of a vector.

[0056]

number

[0057] There are several possible matrices C that satisfy the requirements of Equation (11). For example, if N=5 and t 1n , t 2n is a constant value, that is, the light intensity of the signal light Sig incident on each photodetector element 19 is equal, the following equation (12) can be given as an example.

[0058]

number

[0059] The current component I corresponding to the beat component between the signal light and the local light shown in Eq. (7) Bn is the current component I BnSince the two terms are complex conjugate scalars, it can be expressed as in equation (13). From equation (13), the signal light Sig can be expressed as the Jones matrix J n and transform the local light Lo into the Jones matrix K n In the case where "K" is applied to only the signal light Sig, n † J n ” is applied to only the local light Lo as in the first embodiment. n † K n Therefore, the transformed Jones matrix J n and the transformed Jones matrix K n The transformed Jones matrix H n is defined as equation (14).

[0060]

number

[0061] The vector A of the signal light Sig, which is a complex vector, is converted into a four-dimensional real vector A IQ When expressed as a component vector I B and current component I Bn can be expressed as in equations (15) and (16). n " is the nth component of the vector in parentheses. Also, P is a vector (row vector) P ·n It is an N-by-4 matrix of

[0062]

number

[0063] From equations (8) and (15), equation (17) is obtained. From equation (17), a four-dimensional real vector A IQ That is, it is clear that the complex amplitudes of the X-polarized wave and the Y-polarized wave of the signal light Sig can be decoded by linear calculation.

[0064]

number

[0065] From equation (17), A IQ To obtain (1), the matrix CP must have an inverse matrix, and for the 4 × 4 matrix CP to have an inverse matrix, the ranks (orders) of C and P must be 4. Due to the constraints of equation (11), the rank of C is generally "N-2", so N≧6 is required to correctly decode the signal light Sig. However, when "T1 = αT2" holds with α as a constant, the rank of C is "N-1", so the signal light Sig can be decoded using "N=5", i.e., five photodetector elements 19.

[0066] Hereinafter, it is assumed that the signal light Sig and the local light Lo (in this example, the converted local light) are incident with the same light intensity on each photodetector element 19. The four-dimensional real vector space with the real and imaginary parts of the Jones vector as its axes is defined as IQ space, and the N vectors P of the matrix P are ·n form vertices in the IQ space. As can be seen from equation (13), homodyne detection measures the projection component of the signal light Sig (vector A) onto the local light Lo (vector B). Also, the vector P ·n It can be said that the projection of the signal light Sig onto the optical fiber is being measured.

[0067] In order to obtain the complex amplitudes of the X and Y polarizations of the signal light Sig independently of the local light polarization state, that is, regardless of the polarization state of the local light Lo, the vector P ·n The positional relationship between the vectors P ·k and vector P ·h Inner product of (k≠h) <P ·k ,P ·h > does not depend on vector B. From equation (16a), the inner product <P ·k ,P ·h > is expressed as in equation (18).

[0068]

number

[0069] Here, "H" in the above formula (18) h † H k " is a unitary matrix, but if it is a special unitary matrix (determinant is "1"), then equation (18) can be transformed into equation (19) using complex numbers α and β. Note that α and β are "|α| 2 +|β| 2 =1".

[0070]

number

[0071] From equation (19), "H h † H k If " is a special unitary matrix, then the inner product <P ·k ,P ·h > does not depend on the vector B, so it can be considered polarization independent. h † H k )=det(H h ) -1 ·det(H k ) and this holds true for all different k and h, so it can be seen that local optical polarization independence is achieved by satisfying the above conditional formula (I).

[0072] The index of the measurement error that determines the receiver sensitivity can be expressed using, for example, the variance of the received symbol. If this index is ε, then the index ε is expressed as follows: IQ Using the above-mentioned matrices C and P, it can be expressed as in equation (20). Furthermore, when the light intensity of the local light Lo is sufficiently strong and the local light Lo is equally distributed, that is, when the light intensity of the local light (converted local light) incident on each photodetector element 19 is equal, the reciprocal G of the receiving sensitivity (hereinafter referred to as index G) can be derived as in equation (21) based on equation (18). IQis a four-dimensional real vector representing the measured value, and ΔI is the measured photocurrent I as shown in equation (20a). n Shot noise current I Noisen In the formula, q is the elementary charge, and W is the receiving bandwidth. "<>" indicates the expected value, and "|| || F " denotes the Frobenius norm.

[0073]

number

[0074] It is derived that the index G in equation (21) is uniquely determined when N=5 or 6, regardless of how matrix C is chosen. On the other hand, when N≧7, index G varies depending on how matrix C is chosen, and there are multiple ways to choose matrix C that minimizes index G. This means that the generated interference light and photodetector element 19 are redundant, and there is room for selection of interference light and photodetector element 19 used for coherent reception. In a system with redundancy in which the receiving sensitivity varies depending on how matrix C is chosen, it is also possible to use a configuration in which the way the photocurrent difference is taken is controlled so as to dynamically control matrix C using DSP (digital signal processing) or the like.

[0075] When the model described above was applied to existing coherent receivers, the index G was "8" for the 90° optical hybrid (N=8) and the 120° optical hybrid (N=6), and the index G was "16" for the 72° optical hybrid (N=5). However, when minimizing the index ε as a mathematical optimization problem, no index ε was obtained that would make the index G smaller than that applied to the existing coherent receivers as described above.

[0076] From the above, it is considered that the index G has a lower limit depending on N as shown in equation (22). Therefore, the transformed Jones matrix H is calculated to minimize the index G given by equation (21). nThe theoretical maximum receiving sensitivity can be determined by determining the maximum receiving sensitivity. In the following explanation, the receiving sensitivity will be normalized as a ratio to the receiving sensitivity (theoretical upper limit) of a 90° optical hybrid. The theoretical maximum receiving sensitivity when five photodetector elements 19 are used is 1 / 2 of the theoretical upper limit of a 90° optical hybrid, or "0.5."

[0077]

number

[0078] As mentioned above, homodyne detection is the process of independently measuring each element of vector A in the IQ space as a projection component of signal light Sig (vector A) onto local light Lo (vector B). Therefore, when N≧5 or more, the vector P ·n It is known that the maximum receiving sensitivity can be obtained by distributing the vectors P evenly in the IQ space. In this example, when N=5, ·n Regarding <P ·k ,P ·h >(k≠h) is the minimum value, that is, for any combination of k and h <P ·k ,P ·h > is the minimum value. From equation (19), <P ·k ,P ·h > is expressed as the matrix H h † H k Therefore, when N=5, the maximum receiver sensitivity can be obtained by minimizing the value Rs shown in equation (II).

[0079]

number

[0080] When N=5 is the maximum receiving sensitivity, in the IQ space, each vector P ·nform the vertices of a regular pentacle. Also, the transformed Jones matrix H n is a unitary matrix, and the norm of vector B is 1 (||B||=1), so vector P ·n About "||P ·n ||=1" holds. Therefore, each vector P ·n The regular pentagon formed by N = 5 has its center of gravity (geometric center) at the origin, and its geometric characteristics result in the requirement of the following equation (24): Therefore, when N = 5, the maximum receiving sensitivity is obtained when the value Rs given by equation (II) is -1 / 4 as described above.

[0081]

number

[0082] Fig. 6 shows the results of a numerical simulation of the change in receiver sensitivity due to fluctuations in the polarization state of the local light Lo in the configuration of the coherent receiver 10 shown in Fig. 1. The horizontal axis of the graph in Fig. 6 represents the rotation angle (tilt angle) θ of the polarization of the local light Lo. P and the vertical axis represents the phase shift Δφ of the local light Lo. P The rotation angle θ P is the angle with the X-axis direction as the reference (0), and the phase shift amount Δφ P is the phase amount with the local light Lo in a linearly polarized state as the reference (0). The receiver sensitivity is the normalized value as above.

[0083] Similarly, the results of a numerical simulation of the change in receiving sensitivity due to fluctuations in the polarization state of the local light Lo in a conventional local optical polarization-dependent coherent receiver are shown in Figure 7. Note that "45" in the graph in Figure 7 indicates the point where the local light Lo is a specified linearly polarized wave (rotation angle π / 4, phase shift amount 0), "LHC" indicates the point where the local light Lo is a left-handed circularly polarized wave (rotation angle π / 4, phase shift amount 1 / 2π), and "X" indicates the point where the local light Lo is a linearly polarized wave in the X direction (rotation angle 0, phase shift amount 0).

[0084] As can be seen from the graph in Figure 6, the coherent receiver 10 can achieve half the theoretical upper limit of the 90° optical hybrid sensitivity regardless of how the rotation angle and phase shift amount of the local light Lo vary. In other words, the coherent receiver 10 can achieve the theoretical maximum sensitivity regardless of the polarization state of the local light Lo. Furthermore, the graph in Figure 7 shows that in a local optical polarization-dependent coherent receiver, changes in the phase shift amount do not affect the sensitivity, but changes in the rotation angle significantly affect the sensitivity.

[0085] As described above, it can be seen that the coherent receiver 10 can demodulate the signal light Sig using a reduced number of photodetector elements 19, five, regardless of the polarization state of the local light Lo.

[0086] [Second embodiment] The second embodiment is configured so that signal light and local light are input to the lightwave converter. Note that, except for the following description, the second embodiment is the same as the first embodiment, and the same components are denoted by the same reference numerals and their description will be omitted.

[0087] 8, a coherent receiver 40 in this example generates 5 (=N) interference lights and demodulates the signal light Sig independently of the local optical polarization. The coherent receiver 40 includes a lightwave converter 41, a light-receiving unit 14, and a processing unit 13, which constitute an optical device 11A. The lightwave converter 41 is arranged on the optical path of the signal light Sig and the local light Lo. The lightwave converter 41 is composed of a single transmissive metasurface 42, and the signal light Sig from the optical fiber 17 and the local light Lo from the optical fiber 18, which are arranged on the incident surface side (left side in FIG. 8) of the metasurface 42, are incident on the incident surface.

[0088] The optical fibers 17 and 18 are arranged offset from each other in the in-plane direction (vertical direction in FIG. 8) of the incident surface of the metasurface 42. For example, the centers of the cores of the optical fibers 17 and 18 are spaced apart by approximately 127 μm. The signal light Sig and the local light Lo from the optical fibers 17 and 18 are incident on the same incident region on the incident surface of the metasurface 42 with a predetermined spread.

[0089] The metasurface 42, like that of the first embodiment, has a configuration in which multiple meta-atoms are provided on a transparent substrate. The metasurface 42 converts the signal light Sig incident on the incident region into six converted lights (hereinafter referred to as converted signal lights) that have been subjected to polarization conversion and phase change, and converts the local light Lo incident on the incident region into six converted lights (hereinafter referred to as converted local lights) that have been subjected to polarization conversion and phase change. That is, in this example, when the number of interference lights generated is N, the signal light Sig and the local light Lo are converted into N+1 converted signal lights and converted local lights, respectively. Note that although the metasurface 42 in this example is a transmissive type, a reflective type metasurface can also be used.

[0090] Each converted signal light is emitted from the metasurface 42 so as to be focused at a different position on the light-receiving surface 14a of the light-receiving unit 14. In this example, the converted signal light is focused at positions arranged at equal intervals in a line. Similarly, each converted local light is emitted from the metasurface 42 so as to be focused at positions arranged at equal intervals in a line on the light-receiving surface 14a.

[0091] Where n is 1, 2, . . . , 5, the metasurface 42 focuses the converted signal light and the converted local light on the same line so that the position where the n-th converted signal light is focused coincides with the position where the (n+1)-th converted local light is focused, and emits each converted signal light and each converted local light while shifting them relatively in the direction in which the focusing positions are aligned. This causes the n-th converted signal light and the (n+1)-th converted local light to overlap on the light-receiving surface 14a to generate the n-th interference light, thereby generating five interference lights.

[0092] In this example, the multiple converted local lights and the multiple converted local lights are arranged in a line, but they may be arranged in an arc (on the same circumference), for example, and each converted signal light and each converted local light may be relatively shifted in the circumferential direction as one direction. In this case, a metasurface may be added to control the arrangement of the multiple converted local lights and the multiple converted local lights.

[0093] The optical device 11A configured as described above can be incorporated into, for example, a small ROSA (Receiver Optical SubAssembly), and the coherent receiver 40 can be made compact.

[0094] The metasurface 42 applies a polarization conversion and a phase change to the converted signal lights so that their polarization states differ from each other, and similarly applies a polarization conversion and a phase change to the converted local lights so that their polarization states differ from each other. The metasurface 42 also applies the same polarization conversion and phase change between the n-th converted signal light and the converted local light.

[0095] The light receiving unit 14 is provided with five photodetecting elements 19 corresponding to the five interference light beams to be generated. These photodetecting elements 19 are arranged at positions on the light receiving surface 14a where the first to fifth converted signal light beams are incident, i.e., at light receiving positions. As a result, each photodetecting element 19 receives the corresponding interference light beam. Note that in this example, the sixth converted signal light beam and the first converted local light beam, which are not used for coherent reception, are also incident on the light receiving surface 14a, but they do not have to be incident on the light receiving surface 14a.

[0096] As in the first embodiment, the processing unit 13 obtains the complex amplitude of the X polarization and the complex amplitude of the Y polarization of the signal light Sig from the received light signals of the five interference lights obtained from the light receiving unit .

[0097] The lightwave converter 41 configured as above applies polarization conversion and phase change to each converted signal light and each converted local light so as to satisfy the above-mentioned conditional formula (I). This makes the local light polarization independent. In this example, since N=5, the maximum receiving sensitivity is obtained by setting the value Rs given by formula (II) to the minimum value "-1 / 4".

[0098] Here, m=1, 2, . . . , 6 (=N+1), and the Jones matrix of the transformation of the input light (signal light Sig or local light Lo) by the lightwave converter 41 (metasurface 42) to the m-th incident position on the light-receiving surface 14a is M m Then, the transformed Jones matrix M m On the other hand, when focusing on the light receiving position (interference light) with n being 1, 2, . . . 5 (= N), the Jones matrix J of the signal light Sig to the nth light receiving position is n As shown in equation (25a), n is 1, 2...5 (=N) and "J n =M n ". Furthermore, for the local light Lo, the (n+1)th converted local light is incident on the nth light receiving position. The polarization conversion and phase change given to this (n+1)th converted local light are the same as the polarization conversion and phase change given to the (n+1)th converted signal light. Therefore, the converted Jones matrix K n is expressed as "K n =M n+1 Therefore, the Jones matrix H n is expressed as in equation (26).

[0099]

number

[0100] The function of the metasurface 42 expressed by equation (26) is shown in Figure 9. The metasurface 42 functions as a beam splitter 51, retarders 521 to 526, phase shifters 531 to 536, and focusing lenses 541 to 546 for the signal light Sig and the local light Lo. The same retarder 52 m , phase shifter 53 m and condenser lens 54 m The metasurface 42 emits the sixth converted signal light and the first converted local light, which are not used for decoding.

[0101] The beam splitter 51 splits the incident signal light Sig and local light Lo into six beams. These split beams are polarized and phase-shifted to produce converted signal light and converted local light. m is the angle θ m The optical axis is tilted at , and the phase of the polarization component perpendicular to the optical axis is changed by a phase difference Δφ m This converts the polarization of the m-th converted signal light and converted local light. m is used to shift the phase between the converted signal light and the converted local light, and the phases of the two orthogonal polarization components of the converted signal light and the converted local light are shifted by the same magnitude (phase φ 0m ) to shift. m The converted signal light and the converted local light are focused onto the photodetector element 19 .

[0102] In this example, it is also possible to provide a metasurface for each function, or to realize one function using multiple metasurfaces. Of course, using a single metasurface, as in this example, is preferable in terms of reducing light loss and the number of parts. It is also possible to configure the lightwave converter 41 using optical elements such as beam splitters, retarders, phase shifters, and focusing lenses, and it is also possible to configure it by combining such optical elements with metasurfaces.

[0103] The above transformation Jones matrix H nWhen the above formula is applied to the conditional formula (I), the polarization independence condition is expressed as formula (27). δ is any real number.

[0104]

number

[0105] Here, "2φ 0n +Δφ n = 0”, an example of a solution where Rs = -1 / 4 is shown in equation (28).

[0106]

number

[0107] In the above configuration, we confirmed through numerical analysis that the complex amplitudes of the X-polarized wave and the Y-polarized wave of the signal light Sig can be obtained regardless of the polarization state of the local light Lo, and that the receiving sensitivity is maximized (half the maximum receiving sensitivity of a 90° optical hybrid).

[0108] In the coherent receiver 40 configured as described above, the signal light Sig is demodulated regardless of the polarization state of the local light Lo using five photodetector elements 19, which is fewer than the first embodiment. Furthermore, since the lightwave converter 41 is composed of only one metasurface 42, the number of parts can be reduced, which is advantageous for further miniaturization.

[0109] In the above, the signal light and the local light are incident on the same metasurface to generate the interference light, but the signal light and the local light may be incident on separate metasurfaces, and multiple converted signal lights emitted from one metasurface and multiple converted local lights emitted from the other metasurface may be combined using a half mirror or the like, as in the optical device of the first embodiment, to cause interference at different positions on the light-receiving surface. In this case, for example, one metasurface may be configured to generate the first to fifth converted signal lights, and the other metasurface may be configured to generate the second to sixth converted signal lights.

[0110] Note that the above numbers of converted signal lights and converted local lights converted from signal light and local light are merely examples and are not limited thereto. At least N+1 converted signal lights and converted local lights may be converted. For example, 7 (=N+2) converted signal lights and converted local lights may be generated, and 5 (=N) of the converted signal lights and converted local lights may be caused to interfere with each other to generate 5 (=N) interference lights that are actually used. In this case, for example, the nth converted signal light and the n+2th converted local light may overlap at the light receiving position. Alternatively, the nth converted signal light and the n+1th converted local light may overlap at the light receiving position to generate 6 interference lights, and 5 of the interference lights may be used to demodulate the signal light.

[0111] [Third embodiment] The third embodiment is configured as a multi-channel coherent receiver. Note that, except for the following description, it is the same as the second embodiment, and the same reference numerals are used to designate substantially the same components, and the description thereof will be omitted.

[0112] 10, a coherent receiver 60 includes a lightwave converter 61, a light-receiving unit 62, a processing unit 63, and a divided metasurface 64. The lightwave converter 61 is composed of a single metasurface 65. Six signal light beams Sig1 to Sig6 and local light Lo are input to this coherent receiver 60 from a multicore fiber 66. The multicore fiber 66 has a central core 68 disposed at the center of a cladding 67, and six outer cores 69 disposed at equal intervals on the same circumference centered on the central core 68. In this example, the local light Lo is transmitted through the central core 68, and the signal light beams Sig1 to Sig6 are transmitted by the outer cores 69, respectively.

[0113] The local light Lo and the signal light Sig1 to Sig6 emitted from the end face of the multicore fiber 66 are incident on the incident surface 64a of the segmented metasurface 64, each with a spread. The local light Lo and the signal light Sig1 to Sig6 are incident on the incident surface 64a in the same arrangement as the central core 68 and outer cores 69 at the end face of the multicore fiber 66. The segmented metasurface 64 splits the incident local light Lo into six split local light Loa, and emits the six split local light Loa so that the spacing between them spreads radially.

[0114] The metasurface 65 has six incident regions 71 on its incident surface 65a. The divided metasurface 64 controls the divided local light Loa so that each divided local light Loa is incident on one incident region 71 of the metasurface 65. Furthermore, for each of the signal lights Sig1 to Sig6, the divided metasurface 64 controls the signal lights Sig1 to Sig6 so that each of the signal lights Sig1 to Sig6 is incident on one incident region 71 so that one signal light is incident on one incident region 71.

[0115] The divided metasurface 64 controls the divided local light Loa and signal light Sig1 to Sig6 incident on the metasurface 65 to have the same size as the incident region 71. In Fig. 10, the optical paths of some of the divided local light and signal light are depicted.

[0116] The metasurface 65 is arranged in the propagation direction (to the right in the figure) of the local light Lo and the signal light Sig1 to Sig6, away from the divided metasurface 64. The incident regions 71 are arranged in the same manner as the outer cores 69 at the end face of the multicore fiber 66.

[0117] Each metasurface 65 converts the signal light and split local light Loa incident on each incident region 71 into six converted signal light and six converted local light that have been subjected to polarization conversion and phase change, similar to that incident on the incident region of the metasurface 42 in the second embodiment. Therefore, the converted signal light and converted local light obtained from the signal light and split local light incident on each incident region 71 are focused on the same line, and are emitted from the metasurface 65 with each converted signal light and each converted local light relatively shifted in the direction in which the focusing positions are aligned. As a result, for the signal light and split local light incident on each incident region 71, an nth interference light is generated by overlapping the nth converted signal light and the (n+1)th converted local light on the light receiving surface 62a of the light receiving unit 62.

[0118] In this example, the direction of each line of the converted signal light and the converted local light is radial. Note that the direction of the converted signal light and the converted local light may be any direction as long as the interfering light does not overlap with each other. A metasurface may be added to control the arrangement of multiple converted local light and multiple converted local light.

[0119] A photodetector element 19 is disposed at each position on the photoreceptor surface 62a where the first to fifth converted signal beams corresponding to the signal beams Sig1 to Sig6 are focused. That is, five photodetector elements 19 are provided on the photoreceptor surface 62a to receive five interference beams for each signal beam, and a total of 30 photodetector elements 19 are provided in the photoreceptor unit 62. The processing unit 63 processes the received light signals obtained from the five photodetector elements 19 for each signal beam, thereby obtaining the complex amplitude of the X polarization and the complex amplitude of the Y polarization for each of the six signal beams Sig1 to Sig6.

[0120] In this configuration, the complex amplitudes of the X polarization and the Y polarization can be obtained with maximum reception sensitivity for any signal light, regardless of the polarization state of the local light Lo.

[0121] Although it is compatible with multiple channels, which coherently receives multiple signal lights, the divided metasurface 64 and metasurface 65 are surface-illuminated types, so it is easy to create a two-dimensional array of coherent receivers.

[0122] Although the above describes an example in which six signal lights are input to the lightwave converter, the number of signal lights is not limited to this and may be any plural number. Therefore, the number of signal lights may be any of 2 to 5, or 7 or more. In addition, one or both of the divided metasurface 64 and the metasurface 65 may be reflective.

[0123] In the above embodiments, the case where N=5 has been described, but as mentioned above, N can be an integer equal to or greater than 5. That is, even when N is an integer equal to or greater than 6, local optical polarization independence can be achieved by satisfying the above conditional formula (I).

[0124] If N is an integer equal to or greater than 6, and the local optical polarization independency condition is satisfied, the matrix P is determined to maximize the receiving sensitivity for the polarization of any local optical Lo, and the transformed Jones matrix H n (n is 1, 2, N). For example, the local light Lo is an X-polarized wave as shown in equation (29), and the vector P ·n If we specify as in equation (30), equation (31) is established from the above equation (16a). From equation (31), we can immediately obtain the transformed Jones matrix H n The first column of the components of the h † H k ” is a special unitary matrix, the general form of the two-dimensional special unitary group gives the transformation Jones matrix H n can be determined as shown in equation (32), for example. Since there is a degree of freedom in the 4-row, N-column matrix C that specifies how to take the difference in photocurrent, various transformation matrices H n There is a solution.

[0125]

number

[0126] When N=6, for example, in a configuration in which either the signal light Sig or the local light Lo is converted into converted light, as in the first embodiment, the matrix C can be expressed as shown in equation (33). When this matrix C is used, the converted Jones matrix H n An example of this is given by equation (34).

[0127]

number

[0128] The above transformation Jones matrix H n Following the configuration of the 120° optical hybrid circuit, for the input of local light Lo of X polarization, the vector P .n The matrix P is determined so that each of these forms a vertex of an equilateral triangle. The matrix P for the input of the local light Lo of X polarization is expressed as in equation (35). In the IQ space, each vector P of the matrix P ·n The shape in which each vertex is formed by is called a 3-3 duopyramid or 3-3 fusil. The 120-degree optical hybrid circuit is described in detail in the paper "Po Dong, Chongjin Xie, and Lawrence L. Buhl, "Monolithic polarization diversity coherent receiver based on 120-degree optical hybrids on silicon," Optics Express, Vol. 22, Issue 2, pp. 2119-2125 (2014)."

[0129]

number

[0130] The transformed Jones matrix H expressed by equation (35) nIt was confirmed by numerical analysis that when this is applied, the system becomes independent of local optical polarization and the maximum receiving sensitivity in this configuration is 0.5.

[0131] When N=8, the matrix C can be the same as the method of calculating the difference in photocurrent in a 90° optical hybrid circuit. As in the first embodiment, the matrix C in the configuration in which either the signal light Sig or the local light Lo is converted into converted light can be exemplified by the matrix shown in equation (36). When this matrix C is used, the converted Jones matrix H n An example of this is given by equation (37).

[0132]

number

[0133] The transformed Jones matrix H shown in equation (37) n is a vector P of two points on each axis of the IQ space for the input of local light Lo of X polarization. ·n The matrix P for the input of the local light Lo of X polarization is expressed as in equation (38). At this time, in the IQ space, each vector P of the matrix P ·n The figure whose vertices are formed by is a regular hexadecagon.

[0134]

number

[0135] The transformed Jones matrix H expressed by equation (38) n It was confirmed by numerical analysis that when this is applied, the system becomes independent of local optical polarization and the maximum receiving sensitivity in this configuration is 0.5.

[0136] When configured as above with N = 8, the optical intensity difference between the interfering lights can be extracted as a differential current using a balanced photodetector. In other words, the processing corresponding to the matrix C can be performed in the analog domain, which has the advantage of simplifying the configuration of the coherent receiver.

[0137] In the above, an example in which N≧6 is used and the configuration is the same as that of the first embodiment is shown, but even in a configuration similar to that of the second embodiment, N≧6 can be used. For example, when N=6, the signal light Sig and the local light Lo are converted into seven converted signal lights and seven converted local lights, respectively, and six of these are superimposed to generate six interference lights. [Explanation of symbols]

[0138] 10, 40, 60 Coherent Receiver 11, 11A Optical Devices 13, 63 Processing section 19 Photodetector element 21, 41, 61 Lightwave converter 25, 26, 42, 65 Metasurfaces 64-segment metasurface

Claims

1. Let N be an integer equal to or greater than 5, and let the Jones matrix of the signal light transformation to the nth (n=1, 2, . . . , N) light receiving position be J ｎ , the local light transformation Jones matrix is ​​K ｎ a lightwave converter that converts at least one of the signal light and the local light into at least N converted lights having different polarization states by applying polarization conversion and phase change to the signal light and the local light so as to satisfy conditional formula (I), An optical device, characterized in that the signal light and the local light are made to interfere with each other at N light receiving positions, with at least one of the lights being used as the converted light. [Equation 30]

2. The optical device of claim 1 , wherein the lightwave converter comprises a metasurface.

3. The light wave converter is a plurality of metasurfaces arranged on an optical path of the one light, and only the one light is incident on an incident area on an incident surface and converted into N converted lights, The other of the signal light and the local light is made to interfere with the converted light at the light receiving position.

2. The optical device according to claim 1.

4. The optical device according to claim 3 , wherein the lightwave converter is composed of only two of the metasurfaces.

5. The lightwave converter is composed of one or more metasurfaces, The one or more metasurfaces are: The signal light and the local light emitted from positions shifted in an in-plane direction are incident on the same incident region on the incident surface, the signal light and the local light are each converted into at least N+1 converted lights, and the converted lights of the signal light and the local light are collected so as to be aligned in one direction, and the converted lights of the signal light and the local light are emitted with a relative shift in the one direction so that the N converted lights of the signal light and the N converted lights of the local light overlap at the N light-receiving positions.

2. The optical device according to claim 1.

6. A split metasurface is provided which splits the local light into a plurality of split local lights, The light wave converter has a plurality of incident regions onto which each of the plurality of split local light beams is incident as the local light beam and onto which each of the plurality of different signal light beams is incident, and emits the converted light so that the N light receiving positions corresponding to the plurality of incident regions are different from each other.

6. The optical device according to claim 5.

7. N is 5 and the matrix H ｋ , H ｈ The vector in the first column of (k ≠ h) is H ｋ１ , H ｈ１ 2. The optical device according to claim 1, wherein when Rs is given by the formula (II), the minimum value of Rs is "-1 / 4". [Equation 31]

8. A matrix of 4 rows and N columns that specifies how to obtain the difference between the photocurrents from the plurality of photodetection elements that are arranged at the respective light receiving positions and receive the interference light is C, a Jones vector that indicates the electric field of the local light is B, and a vector that arranges the current components of the signal light in each photocurrent is I. Ｓ , the sensitivity of the photodetector element is R Ｐ , vector H ｎ The first and second components of B are (H ｎ B) １ , (H ｎ B) ２ 2. The optical device according to claim 1, wherein, when N is 5 or more, the index G given by formula (III) is minimized. [Equation 32]

9. 9. The optical device according to claim 1, further comprising a plurality of photodetector elements disposed at the light receiving positions, each of the photodetector elements receiving the interference light.

10. The optical device according to claim 9 ; a processing unit that obtains a complex amplitude of the signal light based on light reception results of the plurality of photodetector elements; A coherent receiver comprising:

Images