JP4040344B2 - Optical network - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a node having a wavelength conversion function or an identification / reproduction function and an optical network for setting an optical path according to an arrangement state of nodes having no function.
[0002]
[Prior art]
In an optical network, each node has a wavelength conversion function and an identification / reproduction function, and converts the wavelength of signal light as necessary, or shapes the waveform of the signal light for transmission. When this node has this wavelength conversion function and identification / reproduction function, the degree of freedom in setting the optical path is increased, but the cost of the entire optical network is increased due to the expensive wavelength conversion apparatus and identification / reproduction apparatus. Become. Therefore, a method for configuring an optical network by using a minimum necessary wavelength conversion device and identification / reproduction device has been studied.
[0003]
[Problems to be solved by the invention]
In order to minimize the wavelength converters and identification / reproduction devices included in each node constituting the optical network, which nodes are equipped with these devices and in what usage state, each of the components constituting the optical network The node needs to recognize it. By doing so, for example, nodes that can be wavelength-converted are known, so that it is possible to perform optimum wavelength selection according to the arrangement.
[0004]
The present invention provides an optical network that notifies a usage status of each function between nodes in an optical network composed of a node having a wavelength conversion function or an identification reproduction function and a node not having the function. Objective.
[0005]
[Means for Solving the Problems]
The present invention relates to a wavelength converter that indicates which wavelength can be converted for each input wavelength to a node having a wavelength converter or an identification regenerator in an optical network having the wavelength converter or the identification regenerator at an arbitrary node. Or a monitoring unit that monitors the usage status of the identification regenerator, and the wavelength converter or the identification regenerator obtained by the monitoring unit of the node that includes the wavelength converter or the identification regenerator at each node constituting the optical network. It is a configuration for sharing usage information.
[0006]
The information obtained by the monitoring unit represents the use status of the wavelength converter or the identification regenerator indicating which wavelength can be converted for each input wavelength by a bit map. Moreover, you may represent with the cost value corresponding to each conversion wavelength for every input wavelength. Moreover, you may represent as a total value or an average value of the cost value for every input wavelength.
[0007]
In addition, the optical network of the present invention reduces the usage status of the wavelength converter or the identification regenerator from a node having a monitoring unit to another node at regular time intervals or when the usage status exceeds or falls below a predetermined reference value. You may make it notify at the time of an initial state.
[0008]
Further, as the usage status of the wavelength converter or the identification regenerator, all or a difference from the previous notification time may be notified.
[0009]
Further, the cost value may be set smaller as the number of unused wavelength converters or identification regenerators increases or the number of unused wavelengths increases. At this time, the cost for using the wavelength converter or identification regenerator at each node and the link usage cost for connecting each node are added up for all candidate routes that can be set between the source and destination nodes, and the minimum cost is calculated. You may make it select the path | route which shows a value.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
(First configuration example of node)
FIG. 1 shows a first configuration example of a node of an optical network according to the present invention.
[0011]
In the figure, a node 10 includes wavelength converters (WC) 11-1 to 11-4 that perform wavelength conversion of signal light of two wavelengths (total of four wavelengths) respectively input from two optical fibers # 1 and # 2. And an optical switch 12 that switches the route and outputs the two optical fibers # 1 and # 2. Here, each wavelength converter has a tunable configuration capable of converting the input signal light into an arbitrary wavelength. The monitoring unit 13 monitors the usage status of each of the wavelength converters 11-1 to 11-4 and stores the obtained usage status in the wavelength converter usage status database 14. The wavelength converter usage status database 14 transmits / receives data of each wavelength converter usage status to / from other nodes.
[0012]
The monitoring unit 13 corresponds to each of the optical fibers # 1 and # 2, and has a table indicating whether or not wavelength conversion from the input wavelength to the output wavelength is possible. In the initial state, all the wavelength conversions from the input wavelengths λ1 and λ2 to the output wavelengths λ1 and λ2 are possible, so “1” is set for both the # 1 table and the # 2 table.
[0013]
Here, for example, when the signal light having the wavelength λ1 of the optical fiber # 1 is converted into the wavelength λ2 by the wavelength converter 11-1, the input wavelength λ1 in the # 1 table is changed to “0” indicating that wavelength conversion is impossible. At this time, the signal light having the wavelength λ2 of the optical fiber # 1 can be wavelength-converted to either of the wavelengths λ1 and λ2, and remains “1”. Further, when the signal light having the wavelength λ2 of the optical fiber # 2 is converted to the wavelength λ1 by the wavelength converter 11-4, the input wavelength λ2 of the # 2 table is changed to “0” indicating that wavelength conversion is impossible. At this time, the signal light having the wavelength λ1 of the optical fiber # 2 can be wavelength-converted to any of the wavelengths λ1 and λ2, and therefore remains “1”.
[0014]
(Second configuration example of the node)
FIG. 2 shows a second configuration example of the node of the optical network of the present invention.
[0015]
In the figure, a node 10 is provided between an optical switch 12 for switching the path of signal light of two wavelengths (total of four wavelengths) respectively input from two optical fibers # 1 and # 2, and an input / output of the optical switch 12. Wavelength converters 11, 15, and 16 that are connected and perform wavelength conversion of signal light are provided. Here, the wavelength converter (WC) 11 has a tunable configuration capable of converting the input signal light into an arbitrary wavelength, and the wavelength converter (λ1WC) 15 has a configuration for performing wavelength conversion to the wavelength λ1. The unit (λ2WC) 16 is configured to perform wavelength conversion to the wavelength λ2. The monitoring unit 13 monitors the usage status of each wavelength converter 11, 15, 16 and stores the obtained usage status in the wavelength converter usage status database 14. The wavelength converter usage status database 14 transmits / receives data of each wavelength converter usage status to / from other nodes.
[0016]
The monitoring unit 13 corresponds to each of the optical fibers # 1 and # 2, and has a table indicating whether or not wavelength conversion from the input wavelength to the output wavelength is possible. In the initial state, all the wavelength conversions from the input wavelengths λ1 and λ2 to the output wavelengths λ1 and λ2 are possible, so “1” is set for both the # 1 table and the # 2 table.
[0017]
Here, for example, when the signal light having the wavelength λ1 of the optical fiber # 1 is connected to the wavelength converter 16 and converted to the wavelength λ2, it can be formally connected to the other wavelength converters 11 and 15, so The input wavelength λ1 in the # 1 table is left as “1” indicating that wavelength conversion is possible (shown by shading in the figure). Further, the signal light having the wavelength λ2 of the optical fiber # 1 can be wavelength-converted to any of the wavelengths λ1 and λ2, and therefore remains “1”. Furthermore, when the signal light having the wavelength λ2 of the optical fiber # 2 is converted into the wavelength λ1 by the tunable wavelength converter 11, only the wavelength converter 15 that performs wavelength conversion to the wavelength λ1 among the three wavelength converters is vacant. Therefore, the input wavelengths λ1 and λ2 in the # 1 table and the # 2 table are changed to “0” indicating that wavelength conversion to the wavelength λ2 is impossible.
[0018]
By the way, the information in the table of the monitoring unit 13 as shown in FIG. 1 and FIG. 2 represents the usage status of the wavelength converter in the node, and by sharing the information with other nodes, the information in the entire optical network is obtained. The usage status of the wavelength converter can be grasped. This makes it possible to efficiently and optimally select wavelengths so as not to select paths that cannot be physically stretched when calculating the path of an optical path, considering the wavelength converter usage status (whether wavelength conversion is possible) at each node. Can be performed.
[0019]
(Format example used for notification of wavelength converter usage)
FIG. 3 shows a format example used for notification of the usage status of the wavelength converter of each node. In the figure, T (type) indicates the type of notification (in this case, the usage status of the wavelength converter), and L (length) indicates the length of the format of the information to be notified. From this point on, the usage status of the wavelength converter is represented by a bitmap. FIG. 1 and FIG. 2 show examples of wavelength conversion to two wavelengths (λ1, λ2). Here, whether or not wavelength conversion to 32 wavelengths is possible is shown. For example, for a channel with wavelength ID = 0, wavelength conversion from 0 to 7 is possible.
[0020]
By the way, the above example expresses the possibility of wavelength conversion by “1” and “0”. Here, for example, in the configuration shown in FIG. 2, when there are a plurality of tunable wavelength converters 11 and fixed wavelength converters 15 and 16, even a channel capable of wavelength conversion by expressing the state as a cost value. Detailed specification is possible, such as preferential use of low cost values.
[0021]
(Example format for displaying cost values)
FIG. 4 shows an example format for displaying cost values. Here, for example, for the channel with the wavelength ID = 0, the wavelength ID = 0 is not accompanied by the wavelength conversion, so the cost is set to “0000”. For example, when the number of wavelength converters from wavelength ID = 0 to wavelength ID = 1, 2, 3 is 10, 5, or 1, the cost is low from “0010” to high according to the number of wavelength converters. The cost is set to “1100”. The same applies to the channel with wavelength ID = 1.
[0022]
(Format example that displays the average cost value for each input wavelength)
FIG. 5 shows a format example for displaying an average value of cost values for each input wavelength. Here, the cost value of each output wavelength is averaged and set for each input wavelength. For example, an average value of “0000” to “1100” is set for the channel with wavelength ID = 0. Alternatively, the total value may be set. Finally, an average value of cost values assigned to all input wavelengths may be added. Thus, by averaging without assigning a cost value for each input / output wavelength, the format to be notified can be reduced, and the load on the network can be reduced.
[0023]
In the configuration described above, the wavelength converter is replaced with an identification regenerator, and the monitoring unit monitors the usage status of each identification regenerator. The same applies to a configuration that includes a wavelength converter and an identification regenerator and that monitors both usage conditions by the monitoring unit.
[0024]
In addition, the timing notified from the monitoring unit of each node is a fixed time interval, when the usage status exceeds or falls below a predetermined reference value, or in an initial state where the wavelength converter or the identification regenerator is not used. At that time, the value may be notified.
[0025]
In the embodiment shown in FIGS. 1 and 2, the wavelength converter usage database 14 transmits / receives data of each wavelength converter usage status to / from other nodes. As shown, an advertising unit 17 may be provided, and the advertising unit 17 may transmit wavelength converter usage data to the next node. Here, a configuration example corresponding to the node shown in FIG. 1 is shown, but the same applies to the node shown in FIG. At this time, all the data in the wavelength converter usage status database 14 may be transmitted, or only the difference in which the usage status has changed from the previous transmission time may be transmitted.
[0026]
Further, as shown in FIG. 7, a centralized control device 20 is provided for each node 10, and the wavelength converter usage status is monitored from the monitoring unit 13 of each node in the wavelength converter usage status database 14 provided in the centralized control device 20. The data may be collected.
[0027]
The present invention shares the information on the use status (whether wavelength conversion is possible, etc.) of the wavelength conversion function / identification reproduction function of each node with the entire network, thereby suppressing the wavelength conversion function / identification reproduction function to the minimum necessary. A specific example of optical network path selection, which reduces network costs, will be described below.
[0028]
First, a method for assigning a cost value to a conversion wavelength at each node will be described. If a wavelength converter is used at a node where the number of unused wavelength converters is small, the remaining resources of that node will be under pressure, so the smaller the number of mounted or unused wavelength converters, the lower the cost. Like that. For example, a method can be used in which the number of wavelength converters mounted and the reciprocal of the unused number are used as cost values. However, when the number of wavelength converters mounted and the number of unused wavelength converters are 0, the cost value becomes infinite (wavelength conversion is impossible). In this case, a predetermined value α other than 0 to 1 is set, It is used for cost value format display and route calculation shown below.
[0029]
An example of calculating the cost value will be described based on the node configuration in FIG. In FIG. 8, the wavelength converter (WC) 11-1 connected to the input port of the node 10 has a tunable configuration capable of wavelength conversion to either wavelength λ1 or λ2. The wavelength converters (WC) 11-2 and 11-3 connected between the input and output ports of the optical switch 12 have a tunable configuration capable of wavelength conversion to any one of the wavelengths λ1, λ2, and λ3. The wavelength converter (λ2WC) 16 connected between the input and output ports of the optical switch 12 is configured to be capable of wavelength conversion to the wavelength λ2. The wavelength converter 11-2 is assumed to be used.
[0030]
The cost value for each conversion wavelength is based on the number of mounted wavelength converters.
λ1: 1/3 (wavelength converters 11-1, 11-2, 11-3)
λ2: 1/4 (four wavelength converters 11-1, 11-2, 11-3, 16)
λ3: 1/2 (two of wavelength converters 11-2 and 11-3)
It becomes. Based on the number of unused wavelength converters (wavelength conversion possibility),
λ1: 1/2 (two of wavelength converters 11-1 and 11-3)
λ2: 1/3 (wavelength converters 11-1, 11-3, 16)
λ3: 1/1 (one of the wavelength converters 11-3)
It becomes.
[0031]
Here, in an optical network in which three wavelengths (λ1, λ2, λ3) are wavelength-multiplexed, assuming that each wavelength network is arranged in parallel, for example, the cost based on the number of unused wavelength converters The value can be expressed as in FIG. Numerical values expressed in fractions indicate cost values for the conversion wavelengths λ1, λ2, and λ3 at the node C.
[0032]
In addition, when the wavelengths of the three signal lights input to the node 10 in FIG. 8 are λ1, λ2, and λ3 and the cost value of each conversion wavelength is considered with respect to the input wavelength, the number of wavelength converters is used as a reference. ,
λ1 → λ2: 1/4
λ1 → λ3: 1/2
λ2 → λ1: 1/2
λ2 → λ3: 1/2
λ3 → λ1: 1/2
λ3 → λ2: 1/3
It becomes. Based on the number of unused wavelength converters (wavelength conversion possibility),
λ1 → λ2: 1/3
λ1 → λ3: 1/1
λ2 → λ1: 1/1
λ2 → λ3: 1/1
λ3 → λ1: 1/1
λ3 → λ2: 1/2
It becomes.
[0033]
Here, in an optical network in which three wavelengths (λ1, λ2, λ3) are wavelength-multiplexed, assuming that the networks of each wavelength are arranged in parallel, for example, input based on the number of unused wavelength converters The cost value of each conversion wavelength with respect to the wavelength can be expressed as shown in FIG. Comparing the cases of FIG. 9 and FIG. 10, for example, as indicated by the underline, the cost value of the conversion wavelength λ2 is ３ regardless of the input wavelength in the case of FIG. 9, but in the case of FIG. It is 1/3 for the input wavelength λ1 and 1/2 for the input wavelength λ3.
[0034]
The wavelength conversion cost obtained as described above can be used for path calculation together with the link cost between nodes, but the number of wavelength converters (and identification regenerators) and the number of wavelengths multiplexed in the optical fiber. Each cost may be weighted according to the size of. For example, if the number of wavelength converters is small compared to the number of wavelengths multiplexed in the optical fiber, the use of such wavelength converters has a large effect on the entire network, so the wavelength conversion cost must be emphasized. Weighing was performed.
[0035]
An example of the format for displaying the cost value is shown in FIG. 4, but the following method is used to display the fractional value as described above. First, the calculated cost value is expressed in floating point (for example, 5.0 × 10 ⁻¹ ), and this is converted into a 32-bit floating point display format. Note that the cost value α indicating that wavelength conversion is impossible is assigned the maximum value among the values that can be expressed in 32-bit floating point representation.
[0036]
When the cost value (0 to 1) is expressed by 4 bits, for example, the cost value is linearly assigned to “0000” to “1110”, and the cost value α indicating that wavelength conversion is impossible is “1111”. Assign to. Alternatively, if the cost value is less than a certain threshold, all are assigned to “0000”, and a cost value that is greater than or equal to the threshold and less than or equal to 1 is assigned linearly to “0000” to “1110” to indicate that wavelength conversion is impossible α Is assigned to "1111".
[0037]
FIG. 11 shows a first example of route selection in an optical network. Here, assuming that two wavelengths (λ1, λ2) are wavelength-multiplexed in the optical network and the networks of the respective wavelengths are arranged in parallel, the wavelength conversion function can be regarded as a route moving between the arrangements. The numerical values in the figure indicate the link cost and the wavelength conversion cost. Here, the lower the remaining resources, the higher the cost. Regarding the wavelength conversion cost, the remaining wavelength conversion resources decrease in the order of node C, node D, node A, and node B. Here, the wavelength conversion costs from wavelength λ1 to wavelength λ2 and from wavelength λ2 to wavelength λ1 are assumed to be equal. The link cost is the same, but the link “1” indicates that there is no remaining resource (used).
[0038]
Under the above conditions, wavelength conversion is required at the node B or the node C in the route from the source node to the destination node. Here, considering only the link cost, route 1 (originating node-node A (λ2) -node B (λ2 / λ1) -node D (λ1) -destination node) is adopted. Including the wavelength conversion cost, path 2 (source node-node A (λ2) -node C (λ2 / λ1) -node D (λ1) -destination node) is adopted. In order to perform such route selection, the monitoring function of the present invention is used.
[0039]
FIG. 12 shows a second example of route selection in the optical network. Here, the use of the identification regenerator at each node is considered to be an action of moving from one link plane to the same node on another link plane, and a cost value is set as one hop. Further, an algorithm is used that counts a route that has passed by using a wavelength converter or an identification regenerator as one hop, and selects a route that minimizes the total cost value from the source node to the destination node. At this time, when the source node is connected to the next hop node by a plurality of optical fibers and wavelengths, one of them is determined as the first hop, and the subsequent route is calculated by the Dijkstra method shown in FIG. Here, w is the total number of wavelengths, and k is the start wavelength. This is repeated for the number of routes to the next hop, and the route with the minimum cost value is selected from the obtained route candidates.
[0040]
In the second example, the cost value of each conversion wavelength is considered individually with respect to the input wavelength. For example, the wavelength conversion cost from the wavelength λ1 to the wavelength λ2 is 0.6, and the wavelength conversion cost from the wavelength λ2 to the wavelength λ1 is 0.3. Other conditions are the same as in the first example.
[0041]
Under the above conditions, the route 1 from the source node to the destination node is the source node (λ1) −node A (λ1, identification regenerator) −node C (λ1) −node D (λ1) − destination node. Its total cost is 1.1. Further, the path 2 is a case where wavelength conversion is performed at the node B. When the source node (λ2) −the node A (λ2) −the node B (λ2 / λ1) −the node D (λ1) −the destination node, the total cost is calculated. Becomes 1.0. Therefore, the route 2 that minimizes the total cost is adopted.
[0042]
【The invention's effect】
As described above, in the optical network of the present invention, since the usage status of the wavelength converter or the identification regenerator in the entire optical network can be grasped, the usage status of each node (whether wavelength conversion is possible) is considered. It is possible to set an optimal optical path efficiently.
[Brief description of the drawings]
FIG. 1 is a diagram showing a first configuration example of a node constituting an optical network of the present invention.
FIG. 2 is a diagram showing a second configuration example of nodes constituting the optical network of the present invention.
FIG. 3 is a diagram illustrating a format example used for notification of a usage status of a wavelength converter of each node.
FIG. 4 is a diagram showing an example format for displaying a cost value.
FIG. 5 is a diagram showing a format example for displaying an average value of cost values for each input wavelength.
FIG. 6 is a diagram showing a configuration example of a node using an advertisement unit.
FIG. 7 is a diagram illustrating a configuration example of a centralized control type.
FIG. 8 is a diagram showing a node configuration used for explaining an example of calculating a cost value.
FIG. 9 is a diagram illustrating cost values based on the number of unused wavelength converters.
FIG. 10 is a diagram showing cost values for input wavelengths based on the number of unused wavelength converters.
FIG. 11 is a diagram showing a first example of route selection in an optical network.
FIG. 12 is a diagram showing a second example of route selection in an optical network.
FIG. 13 is a diagram showing a processing procedure for optical network path selection;
[Explanation of symbols]
10 Node 11 Wavelength converter (WC)
12 Optical Switch 13 Monitoring Unit 14 Wavelength Converter Usage Database 15 Wavelength Converter (λ1WC)
16 Wavelength converter (λ2WC)
17 Advertising section 20 Central control device

Claims (5)

In an optical network having a wavelength converter or identification regenerator at any node,
The node having the wavelength converter or identification regenerator includes a monitoring unit that monitors the usage status of the wavelength converter or identification regenerator indicating which wavelength can be converted for each input wavelength,
In each node constituting the optical network, information indicating the use status of the wavelength converter or identification regenerator obtained by the monitoring unit of the node provided with the wavelength converter or identification regenerator is input for each input wavelength. It is expressed by the cost value corresponding to
The cost value is set smaller as the number of unused or empty wavelengths of the wavelength converter or identification regenerator is larger,
Each of the nodes is shared by the entire optical network by advertising the cost value in the own node to all other nodes constituting the optical network by an advertising unit provided in the own node,
The configuration is such that the cost value for use of the wavelength converter or the identification regenerator at each node is accumulated for all route candidates that can be set between the source node and the destination node, and a route indicating the minimum cost value is selected. An optical network characterized by that . The optical network according to claim 1, wherein the cost value shared by the entire optical network is expressed as a total value or an average value of cost values for each input wavelength. Each node uses the advertising unit provided in its own node to set the cost value in its own node to all other nodes constituting the optical network at regular time intervals or based on a predetermined usage condition. 2. The optical network according to claim 1, wherein the optical network is configured to advertise when a value is exceeded or below, or in an initial state. 4. The optical network according to claim 3, wherein the use status of the wavelength converter or the identification regenerator is configured to advertise all or a difference from the previous advertisement time. The cost value and the link use cost for connecting each node are integrated with respect to all route candidates that can be set between the source node and the destination node, and a route indicating the minimum cost value is selected. The optical network according to claim 1 or 2.

Images