Path setting method for optical switch device

Abstract

The present disclosure is a path setting device for setting inter-terminal paths in an optical switch device, wherein the paths in which leakage due to crosstalk falls within an allowable range is set in the optical switch device.

Description

How to set up a route on an optical switch

[0001] The present disclosure relates to wavelength path services and fiber path services using optical nodes configured with optical switches and the like.

[0002] The All Photonic Network (APN) has been proposed (see, for example, Non-Patent Document 1). Since the APN also incorporates access areas, the input strength for each path differs due to the near-far problem caused by the difference in distance between the terminal and the in-office equipment. Furthermore, the fiber path service described in Non-Patent Document 1 accommodates Distributed Acoustic Sensing (DAS) signals with high peak strength and Quantum Key Distribution (QKD) signals with weak power in the same optical switch.

[0003] However, conventional optical switches and other optical switching devices are designed to accommodate optical signals with similar intensities and allowable leakage. Therefore, when applied to networks that accommodate signals with different intensities and allowable leakage, there is a possibility that inter-path crosstalk may exceed the allowable range of a path in a path that transfers a weak signal. Here, crosstalk often refers to static crosstalk, which is crosstalk that constantly leaks into paths other than the desired path after a route is set. In addition to static crosstalk, there is also dynamic crosstalk, which temporarily leaks into paths other than the desired path when a route is changed. Hereinafter, when simply referred to as "crosstalk," static crosstalk is meant.

[0004] Open All-Photonic Network Functional Architecture, Version 2.0, October 19, 2023. https: / / iowngf. org / wp-content / uploads / formidable / 21 / IOWN-GF-RD-Open_APN_Functional_Architecture-2.0. pdf

[0005] The present disclosure aims to use an optical switch whose maximum crosstalk value exceeds the allowable range of the path, and to enable route setting in the optical switch to be performed within the allowable range of the path.

[0006] In the present disclosure, paths within an optical switching device such as an optical switch are set so that leakage within the optical switching device falls within a tolerance range determined for each path.

[0007] Specifically, the path setting device of the present disclosure is a path setting device that sets paths between terminals (ports) in an optical switch, and sets a path in the optical switch where leakage falls within the allowable range for the path. Here, leakage is mainly due to crosstalk, and a path where leakage due to crosstalk falls within the allowable range for the path is selected from among paths that can connect between terminals.

[0008] The route setting device sets a new route in accordance with at least one of the following policies: (i) Set a new route such that leakage from an existing route falls within the new route's tolerance range; (ii) Set a new route such that leakage from a new path falls within the existing route's tolerance range. Here, leakage is mainly due to crosstalk, and the route within the tolerance range may be set by determining whether leakage from an existing route (path) falls within the new route's tolerance range; or, alternatively, by determining whether leakage from a new route (path) falls within the existing route's tolerance range.

[0009] The path setting device may switch a path with leakage within the path's tolerance. For example, the optical switch may predetermine a first path that sets a path with large leakage. In this case, the path setting device may set a path that is weak against leakage to a path that is a predetermined distance away from the first path. Here, the path is a path where the intersection of the optical switch is a predetermined distance or more. In the case of an optical switch composed of unit switches, the intersection is the unit switch. If crosstalk occurs during connection or input, the intersection is the connection point or input point. If crosstalk occurs during reflection, the intersection is the reflection point. If crosstalk occurs due to diffraction, the intersection is the diffraction point. The predetermined distance may be a path where the number of hops between unit switches (unit switches) constituting the crossbar switch is a predetermined number or more corresponding to the leakage crosstalk of the unit switches, or a path where the distance between the positions of beam incidence in the coupling system, reflection in the reflection system, and diffraction in the diffraction system in a spatial switch is a predetermined distance or more.

[0010] Here, the number of hops refers to the number of unit switches included in a path connecting the unit switches that constitute one path and that pass through them before inputting to the unit switch that constitutes the other path, when multiple switches, for example, multiple unit switches, are connected in a hierarchical structure or the like to form a larger optical switch. For example, if adjacent unit switches have a hop count of 1, then if one unit switch is sandwiched between them in the path, the hop count will be 2. If the unit switch is a 2x2 switch and the inter-path crosstalk within the unit switch is X dB, the crosstalk to paths not connected by the path will be 2X dB for one hop, 3X dB for two hops, and (n+1)X dB for n hops. If crosstalk other than that at the unit switch can be ignored, the crosstalk to paths formed by a given path is the sum of the crosstalk of the unit switches that constitute the path. Therefore, a given distant path is determined based on the hop count of all unit switches that constitute the path, rather than the hop count of one unit switch that constitutes the path.

[0011] In other words, it can be approximated by the sum of the crosstalk between the unit switches that make up a certain path and the unit switches that make up other paths. If both paths share the same unit switch but do not share the path connecting the unit switches, the crosstalk of the shared unit switch itself will be the main crosstalk. When switching one of the unit switches that make up one path, if the signal is directly input to one of the unit switches that make up the other path, the crosstalk will be the extinction ratio (crosstalk) of the unit switch itself plus the extinction ratio (crosstalk) of one stage of the unit switch. When switching one of the unit switches that make up one path, if the signal is directly input to a unit switch that directly inputs to one of the unit switches that make up the other path, the crosstalk will be the extinction ratio (crosstalk) of the unit switch itself plus the extinction ratio (crosstalk) of two stages of the unit switch. If the display is linear, the crosstalk will be the crosstalk multiplied by each other, and if the display is logarithmic, the crosstalk will be the sum of the crosstalks. Here, the number of stages is the number of unit switches that make up the paths connecting the unit switches that make up each path related to crosstalk, and is referred to as the number of hops in this application. Note that if the crosstalk of the unit switches is sufficiently small, the loss in the optical switch is sufficiently small, and there are multiple unit switches with similar numbers of hops, the number of stages can be approximated by the sum of the crosstalk of the unit switches with that number of hops. Note that with a hop count of 0, multiple paths do not share paths between unit switches, so there is only one unit switch with a hop count of 0.

[0012] In a space switch, if crosstalk due to scattering during spatial propagation can be ignored, the main crosstalk is due to leakage caused by beam expansion, especially in the coupling system that converts fiber propagation to spatial propagation or spatial propagation to fiber propagation, leakage caused by beam expansion at the reflection point where the beam propagating through space is reflected, and leakage caused by diffraction at the diffraction point. If crosstalk due to scattering in the fiber-to-space coupling system can be ignored, an optical switcher that does not use reflection, such as a piezoelectric switcher, and inputs the beam directly from the fiber-to-space coupling system to the space-to-fiber coupling system on the opposite side will have crosstalk equivalent to one coupling system. If there are two reflections, a total of three crosstalks, consisting of two reflections and one coupling system, will be superimposed. In other words, just like an optical switcher configured with multiple unit switches, the crosstalk is the sum of the crosstalks along the path. Therefore, a given distance between paths is determined based on the distances at all points along the path, not just the distance at some points along the path.

[0013] Specifically, the route setting method of the present disclosure is a route setting method executed by a route setting device that sets routes between terminals (ports) in an optical switch, and the route setting device sets routes in the optical switch that have an acceptable level of leakage. Here, the setting may be performed based on (the sum of) leakage due to crosstalk (from other paths other than the relevant path to the relevant path or from the relevant path to other paths other than the relevant path) from among routes that can connect ports.

[0014] The route setting device may provisionally place a new route, estimate leakage into the new route or an existing route due to crosstalk in the provisional placement, and select the route if the estimated leakage is equal to or less than an allowable leakage, or redo the provisional placement if it exceeds the allowable leakage.

[0015] A network system according to the present disclosure includes a path setting device according to the present disclosure, and an optical switch configured to set paths for optical signals transmitted and received by terminals in accordance with instructions from the path setting device. The optical switch may have a maximum crosstalk value greater than an allowable leakage defined for the paths of the optical signals.

[0016] The above disclosures can be combined as much as possible.

[0017] According to the present disclosure, by using an optical switch whose maximum crosstalk value exceeds the tolerance range of the path, it is possible to make it possible to set a route in the optical switch within the tolerance range of the leakage of the path.

[0018] 1 shows an example of an APN configuration. 2 shows an example of a system configuration of the present disclosure. 3 shows an example of a route (path) setting method of the present disclosure. 4 shows an example of a route (path) setting method of the present disclosure. 5 shows an example of a system configuration of the present disclosure. 6 shows an example of a route (path) selection in an XB switch. 7 shows an example of dynamic crosstalk in an optical switch. 8 shows an example of a route (path) selection in a spatial switch. 9 shows an example of a route (path) selection in a spatial switch. 10 shows an example of a cascade connection of optical switches. 11 shows an example of a tandem connection of optical switches. 12 shows an example of a configuration in which, in a tandem connection of optical switches, inputs to both of the tandem-connected optical switches can be selected. 13 shows an example of the number of paths that can be accommodated when two types of paths are used.

[0019] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. Note that the present disclosure is not limited to the embodiments shown below. These implementation examples are merely illustrative, and the present disclosure can be implemented in various forms with various modifications and improvements based on the knowledge of those skilled in the art. Note that components with the same reference numerals in this specification and drawings indicate the same components.

[0020] (Configuration of Network System) Fig. 1 shows an example of the configuration of a network system. The network system of the present disclosure is an All Photonic Network (APN) that includes an APN gateway (hereinafter abbreviated as "APN-G") 92, an APN exchange (hereinafter abbreviated as "APN-I") 93, and a controller 91, which function as in-station devices. The APN includes, as in-home devices, an APN transceiver (APN-T), a terminal 83 equipped with an APN-T, and other terminals and termination devices 94. However, a configuration that does not include an APN-I may also be adopted.

[0021] The APN-G92 sets routes connecting the terminals 83 and termination devices 94 that transmit and receive the fiber path signals, which are main signals, and the wavelength path signals, to the opposing devices, and loops back the wavelength path signals, wavelength multiplexed signals, and fiber path signals within its own APN-G92 or forwards them to the opposing devices via the APN-I93 or another APN-G92. The APN-G92 sets routes connecting the terminals 83 and termination devices 94 that transmit and receive control signals to the controller 91, and terminates control signals from the terminals 83 and termination devices 94 within the APN-G92 or forwards them directly to the controller 91, and terminates control signals from the controller 91 within the APN-G92 or forwards them directly to the terminals 83 and termination devices 94. For example, in the configuration of FIG. 1, if the controller 91 and the terminals 83 and termination devices 94 transmit and receive control signals directly, the APN-G92 only forwards them without transmitting or receiving them. For transmission and reception, for example, a fiber cross connect (FXC) 71, which is an optical switch that distributes optical paths as light, a wavelength selective switch (WSS) 72 that multiplexes / demultiplexes optical signals of wavelength paths and switches their paths, an optical amplifier 73, a return unit 74, an arrayed wavelength grating (AWG) that multiplexes / demultiplexes optical signals of wavelength paths, etc. can be used. The AWG can be used as an optical multiplexer (MUX) and an optical demultiplexer (DEMUX), or as part thereof. The AWG and WSS 72 function as a "multiplexer," a "demultiplexer," and a "multiplexer / demultiplexer." In this application, attention is focused on the multiplexing function of a multiplexer and a multiplexer / demultiplexer. For this reason, the terms "multiplexer" or "multiplexer or multiplexer / demultiplexer" will be used to represent "multiplexer," "demultiplexer," and "multiplexer / demultiplexer." The APN-G92 functions as an "optical switch."

[0022] Hereinafter, the functional unit that changes the path in the APN-G will be referred to as an “optical switching device” or “optical switch.” In the following embodiment, the terminal 83 and the termination device 94 have the function of transmitting and receiving main signals, and the terminal 83, the termination device 94, and the controller 91 have the function of transmitting and receiving control signals.

[0023] In this disclosure, the term "optical switch" includes not only FXC, which corresponds to core multiplexing, but also WSS, since similar issues exist in WDM (wavelength division multiplexing)-related WSS. Also, in this disclosure, the term "optical switch" includes optical switches and WSS that switch between multiple cores in a core that performs SDM (spatial division multiplexing) and between modes within a core, which may be used in an APN.

[0024] The optical switch 82 includes a combination of small switches such as a crossbar (hereinafter sometimes abbreviated as XB) switch. For example, an XB switch is a switch in which N unit switches, each of which is made up of 2×2 switches, are connected. ２ The optical switch 82 may be a combination of switches such as a series or parallel degenerate XB switch or a Clos switch. The optical switch 82 may also be part of the functions provided in the WSS. For example, a Broadcast & Select type may use a Select function, a Route & Combine type may use a Route function, and a Route & Select type may use both.

[0025] (Premise of the present disclosure) Crosstalk occurs between paths in an optical switch 82. Crosstalk occurs mainly at intersections where paths are sorted or switched within a path. In an optical switch that combines unit switches, the intersections are unit switches. In an optical switch 82 that uses reflection, the intersections are reflection points. In an optical switch 82 that uses interference, the intersections are interference points. In an optical switch 82 that uses spatial propagation, the intersections are junctions between space and optical fiber.

[0026] The leakage from a path to another path is the value obtained by subtracting the inter-path crosstalk from the strength of the leaky path when expressed logarithmically. When expressed linearly, it is the value obtained by multiplying the strength of the leaky path by the ratio of inter-path crosstalk. The one-to-one inter-path crosstalk value depends on the route of the two paths within the optical switch 82. Inter-path crosstalk is large between paths that are close to each other and small between paths that are far apart. In an optical switch 82 that has reflection points, interference points, junction points, and diffraction points, it depends on the degree of light concentration at those points and the distance between the points. If it is a Gaussian beam, its optical coupling can be expressed by the following equation. (Equation 1) exp(-d ２ / wo２ ) (1) Here, d is the distance from the center corresponding to the distance between points, and wo is the beam waist radius corresponding to the degree of light concentration.

[0027] In this way, optical coupling can be expressed logarithmically as a quadratic function of the distance from the center. In optical switches that use diffraction, the degree of light concentration can be determined from the distance from the center for first-order diffraction, but in the case of higher-order diffraction, it is the distance from the center of each order of diffraction, so the degree of optical coupling can be determined in a similar manner by shifting the center point. In optical switch 82 that combines unit switches such as XB switches, if crosstalk other than that of the unit switches can be ignored, light propagates through each unit switch along the path that connects the unit switches with leakage reduced by the amount of crosstalk.

[0028] Crosstalk includes total crosstalk, which is the sum of crosstalk from all ports other than the path in question. Total crosstalk includes simple inter-path crosstalk as well as crosstalk due to path-to-path propagation, higher-order diffraction, stray light, etc. When light is concentrated in a way that can be approximated by a Gaussian beam, the base of the beam widens but can be quickly ignored. However, when light is concentrated at the edge, it is necessary to take into account the effects of stray light, such as the reflection of light outside the concentrated area. When connecting any port from any port of an NxN switch using an XB switch or the like, if at least one unit switch is shared between paths, crosstalk equivalent to the extinction ratio of the path can flow in from other paths that share the unit switch.

[0029] For example, in an NxN switch, assume that all N ports on the left side have paths connected to any of the N ports on the right side. In this state, consider the total crosstalk of a path that runs from the first port at the top of the left side through the Nth port at the bottom of the right side on the opposite side. Because this path shares at least one unit switch with all other paths, if the propagation loss is negligible compared to the extinction ratio of the unit switch, it will receive crosstalk from the other (N-1) paths equivalent to the extinction ratio of the unit switch. The sum of these crosstalks is the total crosstalk for this path.

[0030] For example, when bidirectional signals flow through a 16x16 switch with an extinction ratio of -40 dB, the one-to-one path crosstalk is -40 dB, and the worst-case total crosstalk (dB) in both directions when light of the same intensity is input to the 32 ports of the 16x16 switch is expressed as follows (Equation 2): Total crosstalk = one-to-one path crosstalk + 10 Log ((number of ports on one side - 1) x 2) = -40 + 10 Log ((16 - 1) x 2) = -25 dB (2) Here, weak crosstalk, such as that propagating through one or more unit switches, is ignored.

[0031] The first term in the formula is the crosstalk per port, i.e., the one-to-one path crosstalk, and the second term indicates that the crosstalk per port is the same and is multiplied by the number of input ports. In other words, (16-1) is the total number of paths minus the number of paths connected to the desired opposite port. The multiplier of 2 assumes that input and output is performed to all 16 x 16 ports, and that the crosstalk is the same regardless of the input and output direction. Therefore, if input is to only 16 ports on one side, or if the crosstalk for input in the opposite direction is negligible, the multiplier of 2 is replaced with 1. If the number of input ports is reduced, the number of ports is reduced by that amount from (16-1).

[0032] The worst case total crosstalk (dB) when only one-way signals are input, or when crosstalk for input in the opposite direction is negligibly small, can be expressed as follows (Equation 3): Total crosstalk = 1-to-1 path crosstalk + 10 Log ((number of ports on one side - 1) x 1) = -40 + 10 L log (16 - 1) x 1) = -28 dB (3) The multiplier of 1 in the latter term in the equation assumes that input is made to all 16 x 16 ports from the same direction, and that the crosstalk is the same in that direction.

[0033] The worst case total crosstalk (dB) when bidirectional signals share only half of the paths in the 16x16 switch is expressed as follows: (Equation 4) Total crosstalk = one-to-one path crosstalk + 10 Log((number of ports on one side - 1) / 2 x 2) = -40 + 10 Log(16 - 1)) = -28 dB (4) In the equation, (number of ports on one side - 1) / 2 indicates that the number of paths input from both ports is half.

[0034] The effect of leakage is influenced by the difference in signal strength between each path. For a path with high strength, the strength of leakage from a path with low strength is relatively small compared to the strength of its own signal. For a path with low strength, the strength of leakage from a path with high strength is relatively large compared to the strength of its own signal. For this reason, it is not possible to determine whether or not the signal falls within the acceptable range based on crosstalk alone. Therefore, in this disclosure, paths with high leakage and paths vulnerable to leakage do not share switches and are set on predetermined, separate routes.

[0035] 2 shows an example of a system configuration according to the present disclosure. The optical switch 82 is an optical switch (optical switching device) provided in an APN such as an FXC 71, and can be any device capable of setting a path for an optical signal and setting the route of the set path.

[0036] A path between an optical switch, an arbitrary device capable of setting a route for a set path, or terminals is set by connecting the connection terminals (terminals, ports) of the optical switch or the arbitrary device capable of setting a route for a set path. In this application, the following actions are referred to as connection (of connection terminals (terminals, ports)), connection change (of connection terminals (terminals, ports)), route (path) setting, route (path) change, or route (path) switching: connecting the connection terminals (terminals, ports) of the optical switch or the arbitrary device capable of setting a route for a set path from a state in which the connection terminals (terminals, ports) are not connected and no route (path) is set, and changing (switching) the connection between already connected connection terminals (terminals, ports) to change (switch) the set route (path).

[0037] The allowable leakage information for each path and the leakage information for each path are input to the route setter 81. Based on this input information, the route setter 81 sets routes between ports in the optical switch 82. For example, it sets a path connecting port P1 and port P3 of the optical switch 82. The optical switch 82 sets a route connecting port P1 and port P3 in accordance with the control signal input from the route setter 81.

[0038] The route setting method disclosed herein is a route setting method executed by a route setter 81, which sets, from among routes that can connect ports, routes in which leakage due to crosstalk falls within the path's tolerable range, to the optical switch 82. Here, crosstalk is, for example, the sum of leakage into a new path that occurs due to leakage from paths other than the new path within the route of the new path when the new path is set, or the sum of leakage from the new path that occurs additionally from the route of the new path to paths other than the new path and leakage from existing paths other than paths that were previously subject to leakage before the new path was added.

[0039] In this case, the route setting unit 81 determines whether the leakage into the new path is within the tolerance range of the new path, and sets the new path to a route where the leakage falls within the tolerance range of the new path. Alternatively, the route setting unit 81 sets the new path to a route where the leakage into one or all existing paths added by the leakage from the new path falls within the tolerance range of each of the existing paths. Alternatively, the route setting unit 81 sets the new path to a route where both are compatible. The determination may be made based on the leakage of the single path if the leakage of the single path is dominant, or may be made based on the sum of the leakage of all paths other than the new path if the leakage of all paths other than the new path is influential. The route setting unit 81 may determine whether the leakage from the new path is within the tolerance range of some or all other paths, and set the new path to a route where the leakage falls within the tolerance range of the other paths. If the leakage from the new path is dominant, the determination may be made based on the additional leakage caused by the new path. If the leakage from paths other than the new path is influential, the determination may be made based on the sum of the additional leakage and the leakage from paths other than the new path to the other paths. Also, a path may be set in which both leakage to the new path and leakage from the new path are within the allowable range.

[0040] 3 shows an example of a route setting method according to the present disclosure. The route setting unit 81 tentatively places paths (S11) and estimates leakage from existing paths in the placement or paths to be installed in addition to existing paths, i.e., the combination of input strength and crosstalk value of the path (S12). If the estimated value is equal to or less than the allowable leakage (Yes in S12), the route is selected. If it exceeds the allowable leakage (No in S12), the process returns to the tentative placement of step S11.

[0041] In step S11, the route setting unit 81 selects one of the routes that can be set for the optical signal. This temporarily places the path. In step S12, the route setting unit 81 calculates a leakage value from other paths to the route selected in step S11, and determines whether the leakage value is equal to or less than a predetermined allowable leakage value (S12). If the leakage value is equal to or less than the predetermined allowable leakage value (Yes in S12), the route is adopted. If the leakage value exceeds the predetermined allowable leakage value (No in S12), the process returns to step S11, and another route is selected from the routes that can be set for the optical signal.

[0042] However, it is also possible to adopt a configuration in which only design is performed in the temporary placement stage of step S11, and no port connection changes (route changes) are made. In this configuration, it is desirable to make port connection changes (route switching) after step S12 becomes Yes and the temporary placement route is adopted. This is suitable when leakage can be predicted without actual placement, because temporary leakage due to temporary placement can also be prevented. If leakage cannot be predicted, the route can be changed at the temporary placement stage, and step S12 can be performed based on actual leakage measurements, etc. In this case, there is an effect of suppressing temporary leakage.

[0043] The temporary placement in step S11 may be designed without actually changing the port connections (route switching), and step S12 may be determined based on the predicted leakage value. If step S12 returns Yes and the actual leakage measured after changing the port connections (route switching) using the temporary placement route exceeds the allowable range, the temporary placement in step S11 may be returned to. In this case, the first step S12 is performed based on the prediction, followed by the port connection change (route switching), and then the second step S12 is performed based on the actual measurement. If either the first or second step S12 returns No, the temporary placement in step S11 is returned to.

[0044] The leakage in step S12 is obtained by subtracting the crosstalk value from the leaky path to the leaky path if the value is expressed logarithmically, or by dividing it if the value is expressed linearly, from the input intensity to the leaky path. By repeating steps S11 and S12, the route setting unit 81 can set a route in the optical switch 82 in which leakage from other paths is within the allowable range for that path. The allowable range may be determined by the leakage into a new path or its sum, or the leakage from a new path into an existing path or the sum of leakage from other existing paths.

[0045] In the present disclosure, a route within the allowable range of the path to be set is set. Therefore, the optical switch 82 may have a maximum crosstalk value greater than the allowable leakage defined for the path of the optical signal. As a result, even if an optical switch 82 is used in which the leakage due to the maximum crosstalk value is greater than the allowable leakage of the path, the present disclosure can keep the leakage into the path within the allowable range. In the present disclosure,

[0046] After step S12, a step of selecting a route with the minimum crosstalk value within the range below the tolerance may be further included. As in the embodiment described below in which the number of accommodated paths is maximized, a route with the maximum crosstalk value within the range below the tolerance may be selected. Unlike conventional methods, this embodiment realizes path setting with a crosstalk value below the tolerance.

[0047] In this embodiment, the route within the optical switch 82 is set based on whether or not the leakage is within the tolerance range of the new path, but the present disclosure is not limited to this. For example, instead of or in addition to the leakage into the new path, the route within the optical switch 82 may be set based on leakage that occurs when a new path is set in the optical switch 82 and the new path leaks from the new path to another path. For example, the route setter 81 determines whether or not the leakage is within the tolerance range of another path different from the new path, and sets the new path to a route in which the leakage falls within the tolerance range of the other path. The same applies to the following embodiments.

[0048] Second Embodiment Fig. 4 shows an example of a route setting method according to the present disclosure. In this embodiment, in step S12, the route setter 81 executes steps S21 and S22. In step S21, the route setter 81 estimates leakage from existing paths in the temporary placement or from paths to be installed in addition to the existing paths. In step S22, if the estimated value is equal to or less than the allowable leakage, the route setter 81 selects the route, and if it exceeds the allowable leakage, the route setter 81 returns to the temporary placement.

[0049] Specifically, the allowable leakage M ｉ and leakage L ｊ→ｉ The inequality (5) is determined based on whether it holds or not. ｉ ≧ΣL ｊ→ｉ (5) Here, the allowable leakage M ｉ is the allowable leakage of path i, leakage L ｊ→ｉ is the leakage from path j to path i, where j≠i.

[0050] Leakage L ｊ→ｉ can be calculated using the following equation: (Equation 6) L ｊ→ｉ =I ｊ xXT ｊ→ｉ (6) where signal strength I ｊ is the signal strength of path j, and XT ｊ→ｉ is the crosstalk value from path j to path i, where j≠i.

[0051] Crosstalk value XT ｊ→ｉIn addition to the direct crosstalk from path j to path i, XT ｊ→ｉ It is also possible to calculate the amount not included in the above and add it separately.

[0052] After step S22, a step of selecting a route with minimum leakage may be further included. As in the embodiment example described below in which the number of accommodated paths is maximized, a route with a maximum crosstalk value below the tolerance may be selected. In this example, the optical signal is formulated as a one-way leakage of the path. In the case of bidirectional leakage, XT is used. ｊ→ｉ Instead of XT ｊ→ｉ，ｄ Here, d means direction and takes two values. The sum is j≠i only when the values ​​of d are the same, and j=i is also added when the values ​​of d are different. Unlike conventional methods, this embodiment realizes path setting below the allowable leakage.

[0053] Third Embodiment Fig. 5 shows an example of a system configuration according to the present disclosure. This embodiment includes a table 83 that stores values ​​of crosstalk between paths obtained by actual measurements or the like. In step S12, the route setter 81 refers to the table 83 and calculates leakage values ​​from other paths using the values ​​in the table 83. In this embodiment, the leakage values ​​can be calculated using values ​​of crosstalk between paths obtained in advance. Therefore, particularly when actual measurements are used, by using the input strength to each path based on the actual measurements, it is possible to more easily set paths that provide the allowable leakage values ​​for each path more accurately than when design values ​​or average values ​​for each path are used.

[0054] Fourth Embodiment In this embodiment, a first route for setting a path with large leakage is determined in advance in the optical switch 82. The route setting unit 81 sets a route that is weak against leakage to a route that is a predetermined distance away from the first route.

[0055] Figure 6 shows an example of path selection in an XB switch. Circles indicate unit switches that make up the XB switch, and dotted lines indicate paths connecting them. Unit switches U11 to U88 are 2x2 switches. Unit switches U11 to U48 are selected as paths with small leakage and allowable leakage, while unit switches U61 to U88 are selected as paths with large leakage and allowable leakage. In other words, unit switches U61 to U88 correspond to the "first path."

[0056] The path on the first path that is closest to the second path is shown by the thick solid line and thick dashed line in a square without corners that surrounds the unit switch used on the first path, and an example of the path on the second path that is closest to the first path is shown by the thick solid line in a square without corners that surrounds the unit switch used on the second path. As shown by the arrows in the figure, the number of hops from the unit switch used on the path on the first path that is closest to the second path to the unit switch used on the path on the second path that is closest to the first path is two. If we assume that the unit switches are MZ switches or the like, their extinction ratio becomes the crosstalk value, and crosstalk is reduced by the extinction ratio for each hop.

[0057] Here, let us assume that the crosstalk of the unit switches is -40 dB, and that the crosstalk between the unit switches used in the first path and the unit switches used in the second path is -100 dB or less, which falls within the allowable leakage range. In this case, the crosstalk from unit switches U61 to U68, which are located closest to unit switches U11 to U48 among unit switches U61 to U88, to unit switches U41 to U48, which are located closest to unit switches U61 to U88 among unit switches U11 to U48, should be -100 dB. If the crosstalk in the unit switches is -40 dB, the crosstalk propagated will decrease by -40 dB for each unit switch hop. For this reason, in order to maximize the number of selectable paths for each of the two paths within a range where unit switches are two or more hops apart, the path setter 81 selects path P, which is the minimum distance of two hops from the first path in the figure. Ｌ１ and P Ｓ１ Set.

[0058] In the example shown, there are three two-hop routes per path, which allows for a small path leakage tolerance, as expressed in the following equation: (Mathematical Expression 7) Σ (Leakage due to crosstalk according to the number of hops at unit switch i) = ((-40) + (-40) + (-40) + 10 Log3) < -110 dB (7) Here, the leakage according to the number of hops at each unit switch i is -40 dB per hop, and there are two hops away, so the crosstalk is the crosstalk resulting from the superposition of the extinction ratios of three unit switches: the unit switch itself (0th hop) present on the route of its own path; a unit switch (1st hop) that is not present on the route of its own path or on the route of the crosstalk source or destination path and that has a route that can connect a unit switch present on the route of its own path to a unit switch present on the route of the crosstalk source or destination path; and a unit switch (2nd hop) present on the route of the crosstalk source or destination path; and this value is the same for all unit switches, and since there are three sets, 10 Log 3 was added. The reason why all unit switches were set to be the same is because the propagation loss inside the optical switch made up of unit switches and the loss due to crosstalk in the unit switches are negligible.

[0059] (Fifth embodiment) In a space switch, an input fiber is connected to a coupling system from space to a desired output fiber via a coupling system from space to space. In a space switch that selects an output fiber from an input fiber based on the incident angle into space, etc., and directly selects an input fiber to an output fiber based on the incident angle from space, etc., crosstalk can usually be ignored in space, but crosstalk at the connection point becomes a concern. When reflection or diffraction occurs between the coupling system into space and the coupling system from space, the sum of crosstalk at each point becomes a concern.

[0060] An example of dynamic crosstalk in an optical switch is shown with reference to Figure 7. Dynamic crosstalk is leakage that occurs when paths are switched between connection terminals of an optical switch, i.e., when the connection of the connection terminals is changed. The change in the connection between the connection terminals of an optical switch is caused by path switching between the connection terminals.

[0061] Examples of leakage due to dynamic crosstalk include the following: (i) For example, leakage occurring when a connection between connection terminals is changed from a connection terminal that is the target of connection change to a connection terminal other than the connection terminal that is the target of connection change. In other words, leakage occurring when a path that connects and constitutes the connection terminal that is the target of path switching is switched to a path that is not the target of path switching and that connects and constitutes a connection terminal other than the connection terminal that is the target of path switching. (ii) For example, leakage occurring when a connection between connection terminals is changed from a connection terminal other than the connection terminal that is the target of connection change to a connection terminal that is the target of connection change. In other words, leakage occurring when a path that is not the target of path switching and that connects and constitutes a connection terminal other than the connection terminal that is the target of path switching is switched to a path that connects and constitutes a connection terminal that is the target of path switching. Note that connection terminals are also called "terminals" or "ports," and paths are also called "paths."

[0062] This embodiment shows a 5x5 schematic diagram (excerpted from https: / / www.fiberlabs.co.jp / tech-explan / about-switch / ) of a MEMS optical switch, which is an optical switch that uses micro-fabricated micromirrors. This switch changes the path by collimating light emitted from input and output optical fibers Fi1, Fi2, Fi3, Fi4, and Fi5 using a lens, reflecting the light off two MEMS mirrors M1 and M2, which can change the angle, and making the light incident on the target input and output optical fibers Fo1, Fo2, Fo3, Fo4, and Fo5.

[0063] The thick lines indicate the selected paths. The free-space optical system includes M1 and M2 that couple (connect) input and output light from the fiber to mirrors with negligible diffusion. M1 and M2 are mirror groups with mirrors corresponding to each fiber. Each mirror of M1 couples (connects) the input and output of the corresponding fiber to a predetermined mirror of M2, and each mirror of M2 couples (connects) the input and output of the corresponding fiber to a predetermined mirror of M1. The input and output fibers may be connection terminals (ports) that couple (connect) the fibers.

[0064] 7(a) to the output optical fiber Fo1 shown in Fig. 7(b), the light from the MEMS mirror M1 is reflected by reflection points P2, P3, P4, etc. that make up the mirror group of the MEMS mirror M2 on the trajectory scanned by the light beam from the fiber, and the light leaks to the output optical fibers Fo2, Fo3, and Fo4 that are not the target for connection. This is dynamic crosstalk.

[0065] Static crosstalk is the sum of at least one of the following: (i) Constant leakage into undesired mirrors near the reflection point due to beam expansion at MEMS mirror M1 and MEMS mirror M2. For example, in the case of MEMS mirror M2, if the desired mirror is P2, constant leakage occurs because the beam incident on P2 expands, causing the beam's tail to overlap with undesired mirrors P1 and P3. (ii) Leakage into undesired fibers at the coupling point (or input point) in the coupling system from the free space system to the fiber. For example, constant leakage for each path occurs because the beam incident on Fo2 expands, causing the beam's tail to overlap with undesired mirrors Fo1 and Fo3 when the desired fiber is Fo2.

[0066] When calculating the sum of leakage, if the leakage due to crosstalk is deemed sufficiently small, the leakage of leakage can be ignored. For example, if the leakage is at -10 dB crosstalk, the leakage of leakage will be smaller than the crosstalk of -20 dB relative to the initial input, and can therefore be ignored. The leakage of leakage of leakage is defined as follows: the source path is the route of input / output fiber Fi5, P1 of MS mirror M2, and input / output fiber Fo5; the destination path is the route of input / output fiber Fi3, P3 to Fo3 of MS mirror M2, and there is an intermediate route of input / output fiber Fi4, and P2 to Fo4 of MS mirror M2. In this case, the leakage of leakage from P1 of the source path to P2 of the intermediate path, and the leakage from the coupling system of Fo4 of the intermediate path to the coupling system of Fo3 of the destination path are considered to be leakage of leakage. For example, the calculations in Equations 1, 2, 3, and 7 are one-time approximations that take into account only leakage, and do not involve second-order approximations or later that include leakage of leakage.

[0067] Figure 8 shows an example of path selection in a spatial switch. The circles in the figure represent intersections and reflection or connection points of the spatial switch. For example, in the case of a spatial switch using a pair of MEMS mirrors, the circles in the figure represent reflection points at the MEMS mirrors of the spatial switch. For example, they are connection points (connection points) that connect a spatial system to a fiber system. For simplicity, we will assume here that only crosstalk at a single reflection point per path is a concern, and crosstalk at other reflection points or connection points can be ignored. When one path passes through multiple intersections, such as multiple reflection points or connection points, the crosstalk at each of these points will be superimposed.

[0068] The path setter 81 selects reflection points P71 to P74, which are located near one corner of a rectangle in which a group of reflection points arranged in a 5x6 array are placed, as paths with large leakage and allowable leakage. From the remaining reflection points P11 to P65, paths with small leakage and allowable leakage are selected. The area around the dashed line centered on reflection points P71 to P74 is set as the crosstalk range of -100 dB.

[0069] At this time, the path setting device 81 selects a path in which crosstalk from a path with large leakage and allowable leakage to a path with small leakage and allowable leakage falls within the allowable leakage of the path with small leakage and allowable leakage. For example, when selecting a path with small leakage and allowable leakage, the path setting device 81 selects a path excluding reflection points P13, P23, P33, P34, and P35 that are adjacent to reflection points P71 to P74 among reflection points P11 to P65. This makes it possible to keep crosstalk from paths with large leakage and allowable leakage to paths with small leakage and allowable leakage at -100 dB or less.

[0070] In this way, by separating the intersections by a predetermined distance, leakage can be kept within a predetermined acceptable range. In the fifth and fourth examples of this example, leakage from a path with large leakage to a path with small acceptable leakage was illustrated, but this combination is not limited to these. Other path combinations can also be separated in a similar manner. For example, if crosstalk between only one pair of paths among multiple paths is an issue, the intersections used by the paths constituting the pair can be separated. This applies, for example, when the same light is branched and both paths pass through an optical switch while propagating a distance shorter than the coherence length after branching, and crosstalk between these paths has a particular impact. The same applies to dynamic crosstalk, regardless of static crosstalk. However, while static crosstalk requires a predetermined distance between the intersections of each path, dynamic crosstalk requires a predetermined distance between the path locus and the intersections near the locus. Here, the locus refers to a path that changes during switching. An intersection near the locus refers to the intersection closest to the path in question. In the switching example of Figure 7, if the mirror on M1 does not change from Fi5, the path that the path passes through will change sequentially from P1 to Fo5, P2 to Fo4, P3 to Fo4, ..., and finally to P5 to Fo1. Therefore, the intersections on the locus are P1, P2, ..., P5 on M2, and Fo5, Fo4, ..., Fo1 of the spatial optical system. For routes that are not subject to switching, for example, routes that go through P2 to Fo4, P3 to Fo3, and P4 to Fo2, the intersections near the locus are the intersections on the locus, and are respectively the reflection point P2 and the connection point Fo4, the reflection point P3 and the connection point Fo3, and the reflection point P4 and the connection point Fo2. The distance between the intersection on the locus and the intersection of other paths at the moment when the path passes through or near the intersection is the target for distance increase. For example, if one of two paths does not switch, the intersection of the path that does not change during the switch will not be closer than a specified distance to the intersection of the path that changes during the switch.If both paths switch, the intersections through which the other path passes will not be closer than a specified distance at each transition.

[0071] In this example, if the route via P1-Fo5 at the intersection on the trajectory is switched without raising the trajectory higher than the route via P2-Fo4, and if the route is not switched to routes via P3-Fo3, P4-Fo2, or P5-Fo1, dynamic crosstalk can be suppressed when switching to a route via P1-Fo5 relative to these routes.

[0072] (Sixth embodiment) Fig. 9 shows an example of route selection in a spatial switch. In Fig. 6 and Fig. 8, the route setter 81 selected routes that are a group of paths with large leakage and allowable leakage and paths with small leakage and allowable leakage. In this embodiment, in the case of a spatial switch, the route setter 81 places ports through which paths with large leakage and allowable leakage pass at reflection points P71 to P74 that are located closest to the ends. The area around the dashed lines centered on reflection points P71 to P74 is the crosstalk range of -100 dB.

[0073] When selecting a path with small leakage and allowable leakage, the path setter 81 selects a path excluding reflection points P11, P21, P22, P13, P24, P25, P51, P52, P61, P54, P55, and P64 that are adjacent to reflection points P71 to P74 among reflection points P11 to P65. This makes it possible to reduce crosstalk from paths with large leakage and allowable leakage to paths with small leakage and allowable leakage to -100 dB or less.

[0074] In this embodiment, a larger area needs to be excluded than in Figures 6 and 8, and the number of reflection points that can select paths with small leakage and small allowable leakage used for other services decreases, and it can be seen that the number of available ports as a whole decreases. This kind of decrease in utilization efficiency is not limited to spatial switches, but also applies when the distance between switches that make up a switch such as an XB switch is increased.

[0075] In contrast, as shown in the fourth and fifth embodiments, regions in the optical switch 82 are allocated according to the magnitude of leakage and allowable leakage. As a result, when changing the route, the signal of a path with small leakage is limited to transition within the range of the small path, and the signal of a path with large leakage is limited to transition within the range of the large path. This makes it possible to prevent large paths from affecting small paths in terms of dynamic crosstalk when changing the route.

[0076] Dynamic crosstalk suppression in the presence of signal light involves changing the path so as to avoid crosstalk during the path change so that the dynamic crosstalk does not have an effect. The latter involves arranging paths that are weak to leakage and have low leakage close to each other as a group. If the dynamic crosstalk between paths that are weak to leakage and have low leakage is within the acceptable range, but the leakage due to dynamic crosstalk between paths with high leakage is unacceptable, this can be avoided by changing the path with high leakage so that the group of paths with low leakage does not pass through the port where dynamic crosstalk occurs when switching. This can be achieved by switching within the group in the case of an MZ, and even with a mirror, it can be achieved by switching within the group. However, this cannot be achieved if the switching operation involves crossing groups and returning to the home position, etc., so the switching operation during the path change is changed within the group. In a method using a mirror, when changing the connecting port (fiber), the path can be changed via a path that is far enough away from mirrors other than the target mirror that crosstalk can be ignored. In this case, it is possible to handle paths that are weak to leakage but have high leakage. Therefore, when changing the distance, the required value of leakage due to dynamic crosstalk can be achieved by selecting a route after the route setting that ensures a distance from an intersection near the trajectory passed during the route change, rather than the distance from the intersection that will be the route after the route setting. If the dynamic crosstalk due to the trajectory during the route change is not an instantaneous value but an average value over a predetermined time, the predetermined value for dynamic crosstalk is relaxed accordingly. For example, if the dynamic crosstalk is a constant value over time t and is an average over time T, it is relaxed to t / T in linear representation.

[0077] When changing the route in Figure 8, paths with large and small leakage are changed so that the optical path does not extend beyond the line connecting the four ports to the left or bottom of the figure. If the path extends beyond the line, for example, by one column or one row, it is sufficient to use only the port with small leakage and weak leakage at -100 dB from the point of extension as the starting point.

[0078] In Figure 9, leakage can be ignored if the ports are separated by one row, so if the trajectory of the optical path when changing the path between paths with large leakage and high leakage resistance passes outside the square with one or more rows of ports in between, i.e., two or more rows removed, no change is made. On the other hand, when changing the path along a trajectory connecting ports of paths with large leakage and high leakage resistance within the square with a straight line parallel to the sides of the square, if dynamic crosstalk cannot be ignored, the effects of dynamic crosstalk can be ignored by not using them as ports of paths with small leakage and low leakage resistance. For example, when changing to P71, P11, P12, P13, and P72, P11, P12, P13, P21, P22, P23, P24, and P25, which are not more than one row away from the trajectory, can be ignored by not using them as ports of paths with small leakage and low leakage resistance.

[0079] If the impact of dynamic crosstalk is small compared to static crosstalk because it is short in time, the dashed range circle will be small. If the dashed circle due to dynamic crosstalk is smaller than the distance between adjacent ports, the impact of dynamic crosstalk can be ignored by not using P71, P11, P12, P13, and P72 as ports for paths that have small leakage and are vulnerable to leakage, rather than P11, P12, P13, P21, P22, P23, P24, and P25.

[0080] Similarly, when ports of paths with large leakage and high resistance to leakage are connected with each other by lines parallel to the sides of the square, static crosstalk causes P11, P21, P22, P13, P24, P25, P51, P52, P61, P54, P55, and P64, while dynamic crosstalk causes P11, P12, P13, P21, P31, P41, P51, P61, P62, P64, P55, P45, P35, and P25 to be ignored by not using them as ports of paths with small leakage and low resistance to leakage. Specifically, the crosstalk effects are P23, P32, P33, P34, P42, P43, P44, and P53. If one moves in a straight line between P71 and P74 and between P73 and P72, the result will be P23, P32, P34, P44, and P53. If one moves in a straight line from P71, P72, P73, and P74 back to the midpoint of P33 and P43 (the center of the square) and then linearly connects to P71, P72, P73, and P74, the ports along that route will become unusable, and conversely, ports P11, P12, P13, P21, P31, P41, P51, P61, P62, P64, P55, P45, P35, and P25, which are closer to the sides of the square, will become usable. Therefore, when dynamic crosstalk cannot be ignored, the port arrangement should be selected taking into account not only the port arrangement but also the trajectory of the beam propagating through space as it changes ports.

[0081] Seventh Embodiment An example of a method for achieving both high-speed switching and low crosstalk is shown with reference to Figures 10 to 12. In Figure 10, the dashed square without corners represents an APN-G, which includes a fiber cross connect (FXC), a wavelength cross connect (WXC), and an add / drop. The WXC is accompanied by an optical amplifier as needed. The add / drop is configured by a combination of a WXC, a wavelength selective switch (WSS), an arrayed waveguide grating (AWG), and an FXC, etc. The add function multiplexes multiple wavelength signals on the access side port into a wavelength-multiplexed signal, and the drop function demultiplexes the wavelength-multiplexed signal into multiple wavelength signals and forwards them to the access side ports.

[0082] The wavelength connection layer uses WXC and add / drop similar to reconfigurable optical add / drop multiplexing (ROADM) to provide wavelength-based end-to-end connectivity, while the fiber path layer uses FXC or similar to provide optical tunnels between nodes regardless of wavelength and forwards light on a port-by-port basis.

[0083] The fiber path layer has the ability to flexibly forward input optical signals or light to different ports regardless of wavelength. Optical switches for fiber-based cross-connects have lower crosstalk levels between connections than wavelength-based switches. Fiber-based cross-connects include optical switches composed of unit switches that are wavelength-independent to the extent that wavelength dependency can be ignored in service provision, space-based optical switches that are wavelength-independent to the extent that wavelength dependency can be ignored in service provision, and robotic patch panels (RPPs) that change paths by changing the fiber ends or the mating of optical waveguides. In general, crosstalk decreases but switching speeds decrease in the following order: optical switches that switch within optical waveguides, optical switches that process beams propagated in space, and optical switches that physically change the mating of the fiber ends themselves or the optical waveguides connected to them. Therefore, it is difficult for many types of optical switches for fiber-based cross-connects to simultaneously accommodate signals such as QKD signals, which have stringent cross-connect requirements, and signals such as distributed acoustic sensing (DAS), which have high optical input levels and, for example, bursty and instantaneous light emission times, with the worst-case crosstalk values.

[0084] In an APN, wavelength path services require high-speed switching in response to user requests, but considering that wavelength multiplexed signals are also used for transmission, the signal strength variation is relatively smaller than that of fiber path services, and crosstalk requirements can be relaxed. In Figure 10, the wavelength connection layer is built on top of the fiber path layer, but it may also be built in parallel with the fiber path layer as in Figures 11 and 12.

[0085] Figure 10 shows a wavelength connection layer built on top of the fiber path layer, which includes an FXC that accommodates both fiber paths and wavelength paths, and an add / drop and WXC similar to a ROADM that accommodates wavelength paths. Fiber path services are transferred via an APN network and another APN-G, or within the APN-G itself. Wavelength path services are transferred to the APN network via the FXC, add / drop, and WXC. Here, the FXC uses a low-crosstalk optical switch such as an RPP to meet the strict crosstalk requirements of fiber paths. Wavelength paths that require high-speed switching are switched using an add / drop with an optical switch that has relaxed crosstalk requirements but can meet high-speed switching requirements, rather than switching using an FXC with low crosstalk but low-speed switching.

[0086] In this example, in a cascade connection example of low crosstalk FXC (fiber path and wavelength path) and high-speed switching add / drop (wavelength path only), RPP or the like, in which crosstalk can be ignored, is used for the FXC, and when switching the wavelength path, the FXC settings are not changed, and only the high-speed switchable add / drop is switched, making it possible to satisfy both the crosstalk requirements of the fiber path service and the switching time requirements of the wavelength path service. However, cascade connection is costly.

[0087] By using this application, it is possible to change the constituent FXC to an optical switch with higher speed switching, and by setting the distance between paths according to the crosstalk requirements, the high-speed switching FXC can be used to switch both fiber paths that require low crosstalk and wavelength paths that require high-speed switching. As a result, if the add / drop in the figure is changed to a WSS consisting of a high-speed switching FXC and an AWG, it is possible to eliminate the optical switching function included in the add / drop and use only the AWG, etc.

[0088] Figure 11 shows an example of a tandem connection between a low-crosstalk FXC 71F and a high-speed switching WSS. While the figure illustrates a combination of an AWG 63 and a high-speed switching FXC 71W as the high-speed switching WSS, other WSSs may also be used. In this configuration, FXCs with different wavelength paths and fiber paths are used depending on the crosstalk and switching speed requirements. In this configuration, the connection terminal (port) of the APN-G connected from the access area connects to an FXC dedicated to the fiber path service and an FXC (or WSS) dedicated to the wavelength path service. Therefore, switching between the fiber path and wavelength path services requires reconnection to a different connection terminal. In other words, the FXC port is fixed, and it is not possible to change which service it is used for.

[0089] By using this application, it is possible to change the constituent FXC to an optical switch with faster switching speeds, and by setting the distance between routes according to the crosstalk requirements, it is possible to use the same FXC instead of different FXCs, so that both wavelength path services and fiber path services can be used from the same connection terminal.

[0090] As shown in Figure 12, by connecting the FXC 71 and the terminal 83 with a 1x2 switch 75, it is possible to arbitrarily select the optical switch (low-crosstalk FXC 71F, high-speed switching FXC 71W) to be connected thereafter, and it becomes possible to select both wavelength paths and fiber paths from the same connection terminal. However, there is a problem in that the cost of providing the 1x2 switch 75 and cascading is high. Furthermore, crosstalk depends on the crosstalk between FXCs. Specifically, crosstalk is the sum of the crosstalks of the 1x2 switch 75 and the FXCs 71 connected to each other in logarithmic terms. Using this invention has the effect of reducing the cost of cascading.

[0091] Eighth Embodiment One of the benefits of blocking is an increase in the number of paths that can be accommodated. Take two types of paths as an example. Figure 13 shows an example of a schematic diagram of the number of paths that can be accommodated when two types of paths are used. In the figure, one is a QKD path as an example of a path with small leakage and allowable leakage, and the other is a DAS as an example of a path with large leakage and allowable leakage.

[0092] N Ｈis the number of ports of the switch when the crosstalk between paths is low enough to allow leakage from high leakage paths to low leakage tolerance paths. When the crosstalk between paths is low enough to allow leakage from high leakage paths to low leakage tolerance paths, the sum of the number of QKD paths and the number of DAS paths is equal to the number of ports. Number of ports N Ｓ As shown in Fig. 1, if the crosstalk between paths is not low enough to allow leakage from a path with high leakage to a path with low leakage tolerance, compensation can be achieved by increasing the path spacing between the path with high leakage and the path with low leakage strength. For example, this can be achieved by increasing the relative distance between the reflection point and the incident point in a spatial system, or by increasing the number of hops of the unit switch in an XB switch.

[0093] However, if the distance between a large leakage path and a small leakage path is increased to compensate for this, the total number of paths will not be sufficient for the number of ports of the switch. Therefore, it is desirable to select a route with the maximum number of paths within a range where the leakage amount of each path satisfies the predetermined value of the allowable leakage amount of each path. For example, if the number of ports is N Ｓ The above arrangement is port number N Ｈ It is chosen to asymptotically approach the above configuration.

[0094] By arranging paths with a large allowable leakage (i.e., paths with a loose tolerance for leakage) closely together and paths with a small allowable leakage (i.e., paths with a tight tolerance for leakage) away from paths with a large leakage, the crosstalk of the entire optical switch 82 does not need to be aligned with paths with a small tolerance for crosstalk. By arranging paths with a small leakage but a loose tolerance as a buffer zone between paths with a tight tolerance for leakage and paths with a large leakage, the accommodation rate can be increased. Furthermore, the number of paths with different tolerances for leakage can be changed for each switch. The accommodation rate can be increased by arranging paths with a small leakage but a tight tolerance for leakage and paths with a large leakage but a loose tolerance for leakage close together. If the crosstalk of paths in the opposite direction is small and the leakage from a path with a large leakage due to that crosstalk to a path with a small tolerance for leakage falls within the allowable leakage, the paths can also be arranged alternately.

[0095] Although the above embodiment shows an example of an APN, the present disclosure can be applied to cases other than APNs where it is desired to avoid the effects of dynamic crosstalk in a network that uses optical switches. Examples include the following: - A network within a data center that includes an optical switch in the path. - A network that performs hitless switching or redundant switching using an optical switch. - In a redundant PON, when changing the PON interface to which some ONUs in communication are connected while leaving the other ONUs in communication. - In a PON, when switching an ONU in communication that was connected to a PON interface other than the PON interface to the PON interface connected to the communicating ONU.

[0096] Ninth Embodiment In the above-described embodiments, a route refers to a path, and a terminal refers to a port. Leakage is mainly caused by crosstalk. Therefore, in the present disclosure, a route is selected from among routes that can connect terminals, such that leakage due to crosstalk falls within the allowable range of that route (path).

[0097] The leakage is mainly caused by crosstalk. Therefore, the route within the allowable range may be set by determining whether the leakage from an existing route (path) is within the allowable range of the new route (path). Alternatively, the leakage from the new route (path) may be set by determining whether the leakage from the new route (path) is within the allowable range of the existing route (path).

[0098] When switching the connection between terminals in the optical switch, the route taken momentarily during the transition of the optical path is set to a locus that does not exceed the allowable route of at least one of the path in question or the other path.

[0099] The path setting device may switch a path where leakage is within the path's tolerance. For example, the following can be given: A path where the intersection of the optical switch is at least a predetermined distance away. Here, the "intersection" refers to the unit switch in the case of an optical switch configured with unit switches. If crosstalk occurs during connection or input, the connection point or input point is the intersection. If crosstalk occurs during reflection, the reflection point is the intersection. If crosstalk occurs due to diffraction, the diffraction point is the intersection.

[0100] Examples of a "predetermined distance path" include a path in which the number of hops between the constituent unit switches (unit switching devices) is equal to or greater than a predetermined number corresponding to the leakage crosstalk of the unit switching devices, and a path in which the distance between the positions of beam incidence in the coupling system of the spatial switching device (spatial switch), reflection in the reflection system, and diffraction in the diffraction system is equal to or greater than a predetermined distance.

[0101] The "predetermined distance" can be determined not from the number of hops of one unit switch that constitutes the path, but from the number of hops of all unit switches that constitute the path. The number of hops is the number of unit switches included in the path connecting the unit switches when a larger optical switch is constructed by connecting multiple switches, for example, multiple unit switches in a hierarchical structure.

[0102] In a space switch, if crosstalk due to scattering during spatial propagation can be ignored, the main crosstalk is due to leakage caused by beam expansion, especially in the coupling system that converts fiber propagation to spatial propagation or spatial propagation to fiber propagation, leakage caused by beam expansion at the reflection point where the beam propagating through space is reflected, and leakage caused by diffraction at the diffraction point. If crosstalk due to scattering in the fiber-to-space coupling system can be ignored, an optical switcher that does not use reflection and inputs the beam directly from the fiber-to-space coupling system to the space-to-fiber coupling system on the opposite side will have crosstalk equivalent to one coupling system. In the case of two reflections, a total of three crosstalks, consisting of two reflections and one coupling system, are superimposed. Therefore, the total crosstalk in the path must be considered, as in the case of an optical switcher configured with multiple unit switches. Therefore, the "predetermined distance" can be determined based on the distance at all points that make up the path, rather than the distance at some points that make up the path.

[0103] (Other Embodiments) By reflecting the path setting method of the present disclosure in the design method of an optical switch, an optical switch with a large maximum crosstalk can be configured. The path setting device 81 of the present disclosure can also be realized by a computer and a program, and the program can be recorded on a recording medium or provided via a network. The program of the present disclosure is a program for causing a computer to realize each function of the path setting device 81 of the present disclosure, and is a program for causing a computer to execute each procedure of the method executed by the path setting device 81 of the present disclosure.

[0104] 1. As shown in Figure 5, the crosstalk between paths may be measured and stored. Alternatively, the crosstalk between paths may be estimated from the switch configuration. 2. Those that are weak to leakage are separated from those with large leakage. 3. By grouping those that are weak to leakage and have small leakage together with those that are strong and have large leakage, the accommodation rate can be increased. 4. Those that are weak to leakage and have large leakage are separated from each other. However, for sparse signals, if it is acceptable as long as the signal conduction times do not overlap, they may be placed closer together depending on the probability of overlap. 5. If there are multiple paths with large leakage but it is probabilistic, the amount of leakage is multiplied by the probability of occurrence, and if it is below a threshold, they are placed near the path that is weak to leakage. Equation (4) can be used to determine whether or not placement is possible. I ｊ When the paths are synchronized in time so as not to overlap, or when it is assumed that they will not overlap stochastically, for example, if there is no time collision up to 10 paths stochastically, then 11 paths can be converted into 2 paths.

[0105] 61: Add / drop 62: WXC 63: AWG 71, 71W, 71F: FXC 72: WSS 73: Optical amplifier 74: Turn-back unit 75: 1x2 switch 81: Route setting device 82: Optical switch 83: Terminal 91: Controller 92: APN-G 93: APN-I 94: Termination device

Claims

1. A path setting device that sets a path between terminals in an optical switch, the path setting device setting a path in the optical switch that brings leakage within the allowable range of the path.

2. The path setting device according to claim 1, wherein the path setting device switches to a path where leakage falls within an allowable range for the path.

3. The route setting device according to claim 1, wherein the route setting device: (i) sets the new route as a route in which leakage from an existing route falls within the tolerance range of the new route, or (ii) sets the new route as a route in which leakage from the new route falls within the tolerance range of the existing route.

4. The path setting device according to claim 1, wherein the optical switch predetermines a first path that sets a path with large leakage, and the path setting device sets a path that is weak against leakage to a path that is a predetermined distance away from the first path.

5. The route setting device according to claim 4, wherein the predetermined distant route is a route in which the number of hops or distance between intersections of optical switches is equal to or greater than a predetermined value.

6. A network system comprising: a route setting device according to any one of claims 1 to 5; and an optical switch that sets routes for optical signals transmitted and received by terminals in accordance with instructions from the route setting device.

7. A route setting method executed by a route setting device that sets routes between terminals in an optical switch, wherein the route setting device sets routes in the optical switch that bring leakage within an acceptable range.

8. The route setting method according to claim 7, wherein the route setting device provisionally places a new route, estimates leakage into the new route or an existing route due to crosstalk in the provisional placement, and selects the route if the estimated leakage is equal to or less than an allowable leakage, or redoes the provisional placement if it exceeds the allowable leakage.

Images