JP2000184408A - Optical network - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a control technology of a large capacity optical network where a communication line can be set by means of dynamic wavelength switching or space switching by planning a communication schedule so that a means whose switching time is short is preferentially used among plural independent space or wavelength switching means. SOLUTION: A node which belongs to an optical network and executes communication is provided with an optical transmission/reception means, two and above independent space or wavelength switching means deciding a space or a wavelength that optical transmitter/receiver user and a control means planning a communication schedule so that a means whose switching time is short is preferentially used in two and above switching means when the node executes a series of communication while an opposite node is changed with time. The node inputs/outputs an optical signal by the optical transmitter 107 connected to four couplers 104-1 to 4 and a space switch 110 and the optical receiver 106 connected to one ring through a wavelength selector 123. The node has a control part 111.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for controlling a large-capacity optical network using wavelength multiplexing or spatial multiplexing and capable of setting a communication path by dynamic wavelength switching or space switching.

[0002]

2. Description of the Related Art In accordance with the progress of computerization, there is a demand for a large-capacity network. In networks using optical communication, research and development to increase the capacity using wavelength multiplexing technology has been underway. However, with the current wavelength multiplexing technology, only about 100 waves can be multiplexed, so it is thought that it will not be possible to immediately respond to the growing demand for information communication.

At present, the performance of a wavelength tunable light source and a wavelength tunable filter is limited. Therefore, if an optical network capable of flexible communication settings is to be realized, the number of available wavelengths is further limited. I will. Therefore, it is conceivable to cope with the increase in capacity required in the future by further multiplexing such a network.

[0004]

FIG. 3 shows an example of a network in which a wavelength division multiplexing network is spatially multiplexed.

In each of the optical rings 103-1 to 103-4 of this network, a plurality of communications can be simultaneously performed on the optical ring by using an optical signal of a plurality of wavelengths simultaneously by using an optical wavelength multiplexing technique. Optical rings having such characteristics are spatially multiplexed to form a multi-ring network. If a logical communication path specified by a combination of an optical ring and a wavelength is referred to as a channel, in this network, each of the nodes 102-1 to 1 selects one from a plurality of channels and performs communication with each other. . Each node includes an optical transceiver having a function of switching a spatially different ring and a function of switching a wavelength to select a channel,
Each time the destination node of communication is changed, the used channel is switched using the optical transceiver.

However, there are some problems in realizing such a network.

The first problem is that the space switching means and the wavelength switching means are completely independent switching means, but the switching time differs. For example, in the case where wavelength switching can be performed many times during spatial switching, the time required for switching relative to the communication time, that is, the utilization efficiency of the optical transceiver, is determined by the time of spatial switching. The advantage of simple wavelength switching is not utilized.

[0008] The second problem is that each node must perform communication setting for deciding a channel to be used with a partner node in order to transfer data. For example, time and processing power are spent for this setting, which is inefficient.

[0009] The third problem is that, with an increase in capacity, communication expended for the communication setting increases, and a plurality of communication setting communication may collide at a node or an improper processing order.

A fourth problem is that each node switches the channel to be used in accordance with the other party each time the destination node of the communication is changed, and a process of selecting an available channel to be the switching destination channel, The processing in the control unit of the node for selecting the used channel correctly according to the other party is heavy.

A fifth problem is that when the networks are interconnected for use in a wide area, it is extremely difficult to obtain a communicable channel due to restrictions on the connection specifications at the connection points.

[0012] These problems are caused by the increase in capacity due to the multiplexing of the optical network. The present invention solves these problems by setting the switching order at the node, selecting the data transfer channel, and using the control information communication path. In the multi-ring configuration, a ring to be used by a receiver is limited, and a solution is achieved by a technique for efficiently performing channel connection in a gateway.

[0013]

According to the present invention, as a first invention for performing efficient node switching, there is provided an optical network for performing communication via a channel designated by space or wavelength or a combination of both. A node belonging to the optical network and performing communication, comprising: at least one optical transmission / reception unit; and two or more independent space / wavelength switching units for determining a space or a wavelength used by the optical transceiver. When performing a series of communications performed while changing the partner node with time, a control means is provided for planning a communication schedule so as to preferentially use a means having a shorter switching time among the two or more switching means.

Further, the switching means is one space switching means and one wavelength switching means, the means having a short switching time is the wavelength switching means, and the control means frequently controls the wavelength switching means. Plan to switch the communication partner node using it.

In addition, the control means frequently uses the priority switching means and plans to switch the communication partner node so as to satisfy a standard for achieving a specific communication quality.

With these functions, the node determines the channel switching order so as to minimize the use of the switching means having the slow switching time while maintaining the required specific communication quality. In addition, the time occupied by channel switching during operation time can be optimized.

According to a second aspect of the present invention for selecting a data transfer channel efficiently and with high reliability, an optical network for performing communication via a channel designated by space or wavelength or a combination of both. Comprises a dedicated control information communication path for exchanging control information between nodes, and a node belonging to the optical network and performing communication comprises: at least one main optical transmitting / receiving means; and the main optical transmitting / receiving means. A space or wavelength switching means for determining the space or wavelength used by the control means and a control means for planning a communication schedule for performing a series of communication performed by the node while changing the partner node with time,
One or more optical transceivers for the control means to exchange control information between nodes via the control information communication path;
The apparatus further includes a unit for transferring the data of the network user via the control information communication channel and a unit for separating / combining the control information.

By providing these functions, control information and user data can be merged and output to the control information communication path or received from the communication path and separated. Safe and efficient use of roads,
The efficiency of the entire network can be improved by using data transfer that does not require communication settings.

As a third invention for simplifying and efficiently using a control information communication path, in an optical network for performing communication via a channel designated by space or wavelength or a combination of both, The network includes a plurality of dedicated control information communication paths for exchanging control information between nodes, and a node belonging to the optical network and performing communication includes one or more main optical transmission / reception means and the main optical transmission / reception means. Space or wavelength switching means for determining the space or wavelength used by the transceiver, control means for planning a communication schedule for performing a series of communication performed by the node while changing the partner node with time, and control information comprising control information A plurality of optical transceivers for exchanging the communication between nodes via the control information communication path, and the plurality of control information communication A function of transmitting data on the communication path so as to average the communication load of each communication path and a function of merging each signal received from the plurality of control information communication paths, assuming that one communication path is equalized. Is provided.

According to this configuration, even when the capacity of the control information communication channel using a plurality of channels is increased, it is not necessary to perform control by identifying the channel, so that control of the communication channel can be kept simple. .

As a fourth invention for improving the efficiency of channel selection processing and the like, in an optical network for performing communication via a channel specified by a combination of both space and wavelength, the optical network is formed in a ring shape. A plurality of connected optical fibers and a node connected to the fiber to input and output an optical signal, the node is via an optical switch having an output connected to each of the plurality of optical fibers, One or more optical transmitters for transmitting an optical signal to a fiber selected from the plurality of optical fibers, and among the plurality of optical fibers, fixedly connected to one fiber determined for each node, One or more optical receivers for receiving an optical signal from the fiber.

According to this configuration, since the ring connected to the receiving node of the communication is fixed, the number of usable channels is reduced, and the process for determining the used channel is made more efficient.

According to a fifth aspect of the present invention for effectively establishing a channel connection at a gateway, in an apparatus for connecting optical networks which communicate with each other via a channel designated by space or wavelength or a combination of both,
The device has an optical wavelength of an input optical signal, or a signal conversion function for changing a combination of an optical transmission line to which the optical signal is input and an optical transmission line to be output, and an output for setting the change. A gateway device, which has a control function of grasping the status of the network connected to the side and determining the change based on the status, wherein the signal conversion function operates according to an instruction of the control function. Used.

[0024] Since the channel setting is performed by the gateway device based on the information of the related network or terminal, an effect is achieved that smooth communication is realized in a state where communication resources are effectively used.

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 shows an example of an optical network to which the invention of claim 1-6 is applied. Node A (102-1) to X (102-24)
Are connected to the network core 101 to form a network. Information is transferred between nodes via this network. Terminals may be connected to each node, routers and switches may be connected, and other networks may be connected. In any case, the information generated by them
It is forwarded to other places via the network shown in 2.

One specific example of the network is:
This is a network constituted by a plurality of sub-networks of optical rings 103-1 to # 4 of # 1 to # 4 in FIG. In each optical ring, a plurality of communications can be simultaneously performed on the optical ring by using an optical signal of a plurality of wavelengths at the same time using an optical wavelength multiplexing technique. Optical rings having such characteristics are spatially multiplexed to form a network. If a logical communication path specified by a combination of an optical ring and a wavelength is called a channel, in this network,
Each node selects one from a plurality of channels and communicates with each other.

Each node has a combination (channel) of a ring and a wavelength assigned in advance, and is provided with an optical receiver for fixedly receiving the channel. Also,
To be able to transmit signals to all nodes in the network,
It also includes a connection ring and an optical transmitter capable of changing the transmission light wavelength. Then, when each node sends data of the first optical frequency of the first optical ring to the first destination node using this optical transmitter, the node switches the optical ring and wavelength and sends the data to the second destination node. To send the next data. As described above, since data is successively transferred to different parties, the combination of a transmitting node and a receiving node dynamically changes.

FIG. 1 shows a configuration example of a node. Node is 4
One optical ring, four couplers 104-1 to 104-4 and space switch 1
An optical transmitter 107 connected via an optical receiver 107 and an optical receiver 1 connected via a wavelength selector 123 to one of the four rings.
06 inputs and outputs optical signals. The input / output of these transceivers and the local input / output interface of the node (not shown) are connected via a routing unit 115. The node has a control unit 111 that manages the optical transceiver, the routing unit, and the like. Further, default channel optical transceivers 108 and 109 mainly for the purpose of information transfer by the control unit 111 are provided with WDM (Wave
It is connected to an optical ring 103-1 via a length division multiplex (coupler) 105-1,2.

The optical transmitter 107 of the above-described node has a wavelength variable light source 121. After the information is modulated by the modulator 122, the optical output is connected to one of the optical rings via the space-switching optical switch 110. This wavelength tunable light source 12
1 and the space switch 110 are controlled by the control unit 111.
The optical receiver includes a wavelength selector 12 for selecting an optical frequency assigned to a node from a ring to which data is transferred.
The optical signal extracted in 3 is received by an optical receiver 106 corresponding to burst reception and converted into an electric signal. after that,
The header processing unit 117 performs header processing and the like, and outputs the extracted information. A node may have a plurality of optical transmitters and optical receivers in order to more effectively use the transmission capacity of the network and facilitate multicast communication. The numbers do not have to be the same.

In addition to the main optical transceiver for data transmission and reception described above, the node also
It includes optical transceivers 108 and 109 for the default channel mainly used for transmission of management information and small-capacity data communication. In the example shown in FIG. 1, the default channel transmitter 109 has a fixed wavelength of λ0 and is connected to the optical ring # 1 (103-1). Similarly, the receiver 108 for the default channel has a reception wavelength of λ0
Fixed and connected to optical ring # 1. In general, the default channel needs to be constantly received in order to grasp the status of the network and respond to communication setting requests from other nodes. Therefore, at least the optical receiver for the default channel is a main optical receiver. Does not double. Also, in order to quickly transfer the data of the communication setting request of the other node to be relayed, it is preferable that the optical transmitter for the default channel does not double as the main optical transmitter. Furthermore, by using this frequency λ0 as a wavelength different from the wavelength used by the main optical transceiver, there is an advantage that a low-cost multiplexer / demultiplexer can be used in the wavelength selector for separating the default channel. For example, the main channel, which is frequently used for an optical amplifier, is set to the 1.55 μm band, while the default channel is set to the 1.3 μm band. In addition, if there are sufficient fiber resources such as a private network, if the fiber for the default channel is made independent, even a multiplexer / demultiplexer becomes unnecessary, and the system is simplified. Note that a channel in which the main optical transceiver communicates with the default channel is also called a bypass channel.

Communication by the bypass channel is, for example,
`` A High Speed Protocol for Bursty T '' by Yoo, M. et al.
raffic in Optical Networks '', (SPIEVol.3230, pp.79-
90, Nov. 1997), using the “tell-and-go” setting method. In this case, each communication is performed according to the following procedure shown in FIG. For example, when the node A transmits data to the node J, the node A first sends an announcement packet 140 that notifies the node J and the intermediate node that the data is to be transferred to the node J. The announcement packet is sent to the default channel of the network. The default channel is a special channel in which all nodes connected to the network can always transmit and receive.

A packet flowing through the default channel is always received by an adjacent node on the ring, and it is checked whether the communication schedule indicated in the notice packet does not disturb other already reserved communication. Make a reservation at that node as needed. After these processes, if the node is not the last node, the notice packet is transferred to the next node. Therefore, each relay node requires a processing time 152. The announcement packet 140 that has left node A
After performing the above-described processing in B, the packet is transmitted as a notice packet 141, and is repeatedly transmitted to the nodes C and D, and finally arrives at the node J.

The node J receives the notice packet 149, determines whether or not the transmission from the node A can be received.
Wait for data from A. The node A unidirectionally sends the burst packet 150 after a certain time t1 (153) after sending the notice packet, unless a packet indicating a notice failure has not returned by that time t1 (153). Accordingly, the burst packet from the node A arrives at the node J after a considerable time has elapsed. In addition, information about t1 is also included in the notice packet, and the timing of arrival of the packet at the node J can be estimated based on the information. According to that,
Node J is ready to receive. For example, in the optical receiver, by setting the reception gain setting and the clock synchronization phase information stored in the past communication from the transmission node A as the initial value of the receiver, the initial pull-in time in the burst reception can be reduced. Try to shorten it. Also, if the necessary buffer area and the routing capability are also secured at this time, smooth burst / packet processing can be realized.

Next, specific examples and effects according to claims 1-4 will be described. In order to perform communication as described above, the node, when transmitting, matches the destination node,
You will always choose a ring and a wavelength. In the bypass communication transmitting section, this selection is performed using two independent switching means of wavelength switching and spatial switching as described above. However, as described above, the first problem is that the time required to switch the respective means is significantly different. The wavelength switching of the optical transmission unit is performed at a speed of the order of microseconds by using a distributed Bragg reflector type semiconductor laser (DBR-LD) as a light source and switching the current applied to the wavelength control electrode. On the other hand, in the spatial switching, a fiber output port is determined by a voltage applied to the piezo element by using a switch that mechanically moves a connection portion of the fiber by a piezo element. The switching speed is on the order of milliseconds. Therefore, once the space switching is performed, the control unit of each node performs communication that can be performed in the ring network as much as possible, and then performs the space switching only after that. By doing so, the frequency of space switching can be reduced and time can be effectively used.

Further, as described above as the second problem, even when the capacity of the data to be transferred is small, it is necessary to perform the optical communication by performing the wavelength switching and the space switching. Since the data transfer time is shorter than the switching time, the communication capacity of the network is wasted. Further, if the data transfer time is shorter than the time required for the communication setting, the processing becomes a burden on the control unit. Therefore, by utilizing the fact that each node has a communication channel for communication setting: a default channel, such short or rare data transfer is performed as follows.
Implement communication via the default channel. By using the function of safely joining / branching a notice packet or a control packet flowing through the default channel with a data packet, the default channel can be shared with reliability.

The scheduling performed by the control unit of each node to determine the order of communication partners takes into consideration the amount of data stored in each node, the service reservation status for each destination, and the load on the destination node. Is done. When a node receives various communication packets from devices connected to the outside of the network or another network, it first determines a destination. Next, it is determined whether the packet is transferred by the main bypass route or the packet transferred by the default channel. When forwarding on the default channel, the packets are temporarily stored in a buffer connected to the default channel transmitter, and the packets in this buffer are
Sent to the default channel sequentially. On the other hand, when transferring by the bypass route, the data is stored in a queue classified for each destination node. In some cases, queues may be classified and stored for each service and each flow within the same destination. These classifications are based on many QoS (Qual
In some cases, the setting is made by a request from the communication user, as in the case of an IP router, and in the case of tag switching used in an IP router, the setting is made voluntarily by a node for the purpose of increasing the efficiency of the network. The activation / termination of these settings can be used as they are by applying implementation according to existing protocols to the routing unit and the control unit. Also, for a node having multiple receivers, the number of queues is independent of the number of receivers. The control unit that performs the scheduling sets the communication settings so that the communication of the bypass channel is performed in order from the counterpart of the buffer in which many of the buffers are stored, based on the accumulation state of the packets in these buffers.

At this time, the control unit of the node having the function according to the first to third aspects of the present invention considers the difference in the switching time for each switching method. Classify by ring. Then, based on the accumulation status of the buffer in which all the destinations for each ring are put together, ring movement scheduling is first performed. Furthermore, scheduling of partner switching, that is, wavelength switching, is performed based on the accumulation status of each partner buffer in each ring. The control unit 111
According to the created scheduling, a notice packet is transmitted to the default channel, and communication setting is sequentially performed. However, when communication is set on a ring, communication is not set with all the other parties on the ring. If the buffer does not store a sufficient amount of packets, for example, packets whose burst length during transmission is longer than the amount that can be compared with the wavelength switching time, transmission to the destination may be postponed. However, even if the destination buffer on the specific ring is full, even if the buffer with the higher priority is transmitted, the communication may be switched to another ring once.

As described above, there is another point in scheduling such that a reasonable service is realized on average. Bandwidth, bandwidth usage efficiency, delay,
Rather than satisfying all of various service factors such as delay fluctuation and packet loss rate, the quality to be realized differs depending on the operation policy of the node or the network. For example, a case where the maximum delay time is an important factor will be described. Here, the control unit schedules communication of the bypass channel such that the maximum delay time received by a packet passing through the node is minimized. For example, "To reduce the number of setting failures, send an advance notice packet with sufficient t1 to the other party with high load, and acquire the communication right periodically,""Elapsed time Even in a buffer in which packets are not accumulated enough to use the bypass channel, the packet in the buffer is transferred to the default channel buffer and sent out via the default channel, "and the rule is applied.

FIG. 5 shows an example of a processing procedure in the control unit 111. From the queue status for each destination, etc., the ring switching order for the next T / 2 hours and the residence time in each ring are determined (step 1).
61). Resources such as necessary communication time are secured for a queue having a higher priority among the queues of the nodes belonging to the ring (step 162). Allocate the remaining ring residence time,
The order of wavelength switching and the residence time at each wavelength are determined from the status of the queue between the nodes belonging to the ring (step 163). The waiting packet in the node row queue to which no wavelength has been assigned in the ring is moved to the default channel transmission buffer (step 164). Subsequently, it is determined whether or not the ring is the last one. If it is not the last ring, the process proceeds to step 162;
Proceed to 66. In the case of the last ring, time zone t0-t0 + T
/ 2 scheduling is completed (step 16
6). Then, by repeating the processes 161 to 166 every time T / 2, the delay that a packet can receive in the queue can be suppressed to T at the maximum. Then, after the process 161 related to the schedule of the space switching, the process 162 to the process 165 related to the schedule of the wavelength switching are repeated by the number of the optical rings, and are configured in an independent two-stage procedure. Is performed.

As described above, the total communication efficiency can be improved by performing the scheduling in consideration of the difference in the switching time in addition to the storage status in the buffer and securing the necessary services. This is the greatest advantage of using the invention.

The notice packet is generated and relayed by the control unit 111. According to the schedule, the use of the bypass channel of the combination of the optical ring and the wavelength determined by the other node is reserved for the partner node and the node in the middle thereof. Figure 6
Shows an example of a notice packet. The notice packet 170 includes, as data, a source address, a destination address, t1 (173) indicating actual data transmission timing, a scheduled burst size (174), and the like. Since the default channel also carries other normal communication packets, the header is used in common. For example, in a network serving IP packets, a 20-byte IP header compatible unit 171 is used. However, the notice packet is highly important information,
After being converted into an electric signal by the default channel receiver 108, the selector 119 is immediately selected from other data and passed to the control unit 111. This is to prevent a packet from being discarded in the worst case when the data passes through the routing function via the buffer as in the case of normal data, incurring a large delay in the case of congestion. Not only the notice packet but also the control packet for exchanging control information between the nodes is handled in the same manner. For this reason, the notice packet and the control packet include a code indicating this in the header. For example, the version of the IP header compatible unit 171 shown in FIG.
In the 4 format, the TOS bit or Protocol field is
In version 6 format, the Priority field is used for that. In this way, by using the priority function of the existing protocol, a selector can be created with a slight improvement to the existing header processing program or element.
After processing the notice packet, if the own node is not the final destination, the control unit relays the notice packet 170 and sends it to the next node. This next node is an adjacent node on the route for reaching the destination node. In the case of a ring network as in this embodiment, there is only one adjacent node regardless of the destination.

In connection with this, a notice packet or a control packet generated by the control unit is transmitted to the optical transmitter 10 for the default channel.
Also when inputting to 9, it is controlled so that it is transmitted with higher priority than normal data coming from the default channel queue.

The control unit 111 also performs control to transfer data from the destination-specific queue to the bypass channel optical transmitter according to the created schedule. First, the content of the queue is transferred to the header processing unit at a timing at which the content of the queue can be burst and transmitted from the optical transmitter at the transmission time notified by the preliminary notice packet. If the node has a retransmission function, the burst may be retransmitted, so the data is also left on the buffer during the transfer. And
The data is deleted from the buffer only when retransmission is not performed within a fixed time t2 after the transfer. In the case of a network that handles IP format packets, the data transferred to the header processing unit 117 is one or more IP packets cascaded (FIG. 7). The header processing unit 117 attaches a header compatible unit 182 for encapsulating this data. In the communication on the bypass channel, the transmitting node and the receiving node already know each other's nodes, and the intermediate node only passes the optical signal as it is. The description of the / receive node address and the data format / protocol of the burst is not required in the header. For burst errors, each of the IP packets 183 to 189, which are data, can be detected, so that only error control of the IP header compatible unit 182 is necessary. Therefore, the field indicating the burst length used to determine the end position of the burst at the time of reception is the most important in the IP header compatible unit 182, and a field obtained by adding a field for error correction to this field is the minimum required. It is a header. This format enables efficient data transfer.

However, it is better to have a source / receiver node address for monitoring optical transmission at an intermediate node.
Because it is possible to set options for future expandability and the default channel exchanges IP packets as is, it is natural to use the IP header even in the bypass channel, in this case the packet format shown in Fig. 7 It is better to use an IP tunnel using 181.
The same applies to networks provided for other packet service systems. In the case of a network in which a plurality of service methods are mixed, such as IP and ATM, a unique header that satisfies the above conditions is required.

The burst to which the header is added is input to the burst optical transmitter 107 for the bypass channel. Therefore, scrambling suitable for the optical transmission path is applied, a preamble used for synchronization pull-in on the receiving side is added, and the preamble is input to the modulator 122 and converted into an optical burst.

When an unacknowledged packet is returned from an intermediate node or a partner node in response to a notice packet, the node attempts to reconfigure or change the destination of the communication node. When the frequency of receiving the non-permission packet is high, the load of the ring is determined to be high, and the ring is moved to another ring. Alternatively, ring loading may be determined by measuring the frequency of unauthorized packets passing through the default channel. In this way, each node
By grasping the load of each ring directly or indirectly,
Give priority to a free ring. However, a notice to a node where communication is concentrated increases the probability of non-permission, and if communication is abandoned each time, communication to the node becomes impossible. Therefore, it is necessary to wait until communication can be repeatedly performed at random, or in accordance with a predetermined rule.

When an unauthorized packet is returned,
Depending on the setting of the delay time t1 from the notice packet,
The burst may have already been sent. At this time, there may be a system configuration in which the data constituting the burst has already been erased from the buffer and the burst cannot be retransmitted. Normally, an upper layer protocol is also provided with an error control mechanism such as retransmission, and an application can operate without retransmission at the optical network level, so this configuration is possible.

There is another problem in transmitting a burst before receiving the unpermitted packet. that is,
A burst that is transmitted despite being disallowed collides with another burst that was allowed to be transmitted (accurately, was not disallowed), and the receiver failed to receive the burst that was granted. . The same situation also occurs when an unauthorized packet is discarded on the way due to an error or the like, or when a loopback by the protection function operates. Therefore, when the burst does not arrive at the time set by the notice packet or the received burst cannot be reproduced, the receiving node transmits the burst to the node which is presumed to have transmitted the burst. It is good to transmit a control packet for notifying the failure of the communication via the default channel. However, since the network is mainly composed of burst communication, retransmission gives a significant delay to data. In order to avoid this, it is better to provide a function to discard unauthorized burst transmission at each node, or to provide a function that prevents an optical signal from being transmitted when an optical signal is already transmitted on a channel.
This function is based on the acousto-optic effect filter (AOTF: Acousto Optic tunabl) which consists of a space switch and a coupler for multiplexing the output of the bypass transmitter to the optical ring.
e filter). AOTF has a function of combining and demultiplexing any wavelength at any degree. By using the AOTF as a multiplexer for a transmission optical signal and simultaneously extracting and monitoring a part of an optical signal of a necessary channel, the above-described countermeasure can be taken in the case of an abnormality.

As described above, knowledge of the communication status of other nodes also contributes to efficient scheduling. Therefore,
The control unit creates and manages a table for grasping the communication status from a notice packet or the like passing through the default channel. In this table, for example, which node communicates with each other node and when is recorded.

The network is designed so that each ring can operate with an average load of 50% or less, for example, in order to reduce the probability that a randomly generated communication reservation will fail due to packet collision. For example, if there are 25 wavelengths that can be used in each ring and the total number of nodes is 100 stations, all the nodes can make full use of the wavelengths and communicate with 4 rings. It is difficult to achieve with the go method. However, if, for example, eight or more rings are provided and receivers are added so that a considerable capacity can be secured at each node, the packet or burst discard rate is sharply improved, and communication can be performed smoothly.

Further, a high-load node, for example, has the effect of lowering the load on each receiver by subordinate to a plurality of rings by a plurality of receivers and at the same time distributing the load evenly to each ring. As a result, it is not necessary to significantly increase the number of rings even by the line setting using the tell-and-go method.

In the above example, the main optical receiver provided in the node uses the invention according to the sixth aspect of the present invention, in which the connected optical ring is fixed and the reception wavelength is also fixed. I was For this reason, as described above, the transmission side node does not need to search for and arrange an available channel when making communication settings, thereby reducing the processing required for channel setting. In particular, since the network handled by the present invention uses a plurality of dimensions such as wavelength and space, there are many combinations of wavelengths and spaces for specifying channels.
For this reason, as pointed out as a fourth problem, there is a problem that the arrangement of usable wavelengths and the processing of the arrangement thereof become heavy. However, as described above, this problem is solved by the present invention. Furthermore, more efficient channel setting processing can be realized by the knowledge about the communication status of the other nodes and the reduction of the receiver load by installing a plurality of receivers.

The switching means used in this embodiment is a combination of a spatial switch for driving a fiber by a piezo element and a transmission wavelength switching by a DBR-LD. Is higher than the frequency of space switching. This can of course be considered in various cases depending on the switching means to be used. For example, transmission wavelength switching by a DBR-LD and a semiconductor optical gate (SOAG: Semi conductor Optical Amplifier)
When an optical transmitting / receiving unit using a spatial switch by er Gate is configured, wavelength switching is performed on the order of microseconds, whereas spatial switching is performed on the order of nanoseconds, which is sufficiently faster than wavelength switching. Also in this case, the effects of the present invention can be obtained by exchanging the usage of the spatial switching and the wavelength switching in the above embodiment. Further, even when a combination of fast switching and slow switching is different for each node, efficient communication is realized by applying the present invention in which fast switching is preferentially used.

As a second embodiment of the present invention, a case where the ring and wavelength of the optical receiver for the bypass channel of each node are variable in the network of FIG. 3 as in the above embodiment will be described below. In this case, the combination of the node transmitting and receiving the signal, the ring used, and the wavelength used dynamically changes.

FIG. 8 shows a configuration example of a node. The difference from the node in the previous embodiment is that the burst optical receiver 106 is connected to all the optical rings via the space switch 110-2 controlled by the control unit 111 and the wavelength selection units 123-1 to 12-4. It is a point. In this node, the burst optical transmitter 107
Has a wavelength tunable light source as a transmission unit for bypass communication for transmitting information, and its output is connected to each ring via a space switching type optical switch 110. Burst optical receiver 10
6 is to select only one optical signal selected by the wavelength selector of each ring to select the optical frequency at which data is transferred.The optical switch selects only one optical signal. Convert. After that, header processing is performed, and the extracted information is output. A node may have a plurality of optical transmitting units and optical receiving units in order to more effectively use the transmission capacity of the network and easily cope with multicast communication. In that case, the number of optical transceivers is not always the same.

The node is provided with an optical transceiver for the default channel mainly used for small-capacity communication and transmission of control information, in addition to the optical transceiver for the bypass channel, as in the above embodiment. is there. Each communication is performed according to the procedure shown in FIG.

The node J that has received the notice packet determines whether or not the transmission from the node A can be received, and if it can receive the packet, adjusts the reception wavelength of the optical receiver for the bypass channel to the wavelength designated by the node A. Then, a burst packet from the node A arrives. Node A sends the burst packet unilaterally at a certain time t1 after transmitting the notice packet, unless a packet indicating the failure of the notice returns by that time t1. If the node J receives the notice packet within t1, the node J can receive the packet from the node A from the beginning.

To perform communication in this manner, the node always selects a ring and a wavelength when transmitting and receiving. In the bypass communication transmitting / receiving section, this selection is performed using two independent switching means of wavelength switching and spatial switching as described above. However, the time required for switching each means is significantly different.

In the case of the optical transmission unit, the wavelength switching is performed at a speed on the order of microseconds. On the other hand, the space switching speed is on the order of milliseconds. Therefore, once the space switching is performed, the control unit of each node performs communication that can be performed in the ring network as much as possible, and then performs the space switching only after that. By doing so, the frequency of space switching can be reduced and time can be effectively used.

In order to realize the above functions, the control unit of each node always displays which node is communicating with which ring, and in particular, so as to grasp as much as possible which optical receiver is used in which ring. Is managing. This table is managed so that the latest status is always reflected by exchanging between the nodes the above-mentioned communication announcement packet flowing through the default channel and the table created by each node based on the packet. Further, if each node broadcasts its existence information to the default channel every time it changes the ring used by its own node, each node can easily grasp the situation, which is even better. The node presence information is sent immediately before and after the transceiver of each node changes the connection ring. The information immediately after the change includes, for example, a scheduled stay time, and is also used by a node that wants to communicate with the node to determine a communication schedule. Then, each node determines the destination of the notice packet based on this table.

However, the notice packet flowing on the default channel is not transmitted to all nodes, and the exchange of tables between other nodes and the presence information flowing from each node are not always transmitted correctly to all nodes. For this reason, the table may not keep track of the status of all nodes or may not reflect the latest status. The control unit manages the table and determines the communication schedule taking this point into account. . For example, although the information obtained from the notice packet is reliable, the information exchanged between the nodes may be out of date, so the table is updated in consideration of this point. Thereby, even when the communication setting based on the table is performed, the success rate of the communication is improved.

As described above, when each node communicates with priority given to wavelength switching, there is a possibility that a partner is not always on the same ring even when trying to communicate, and the node may continue to chase this partner. is there. For such an opponent,
By sending a notice packet by making a notice reservation or by designating both the ring and the frequency, and by ensuring a sufficient space between the notice packet and the data packet, the chasing state can be reduced. Although it is good to announce a communication request fairly early in this way, it is still possible that it has already been decided on which ring to be at the desired communication timing, so it is also possible to obtain the schedule of the other party via the default channel in advance. good.

In the default channel, small-capacity communication is also performed on the default channel. Therefore, if the number of nodes increases, the capacity of one channel becomes insufficient. Default channel is low load,
If the operation is not performed with a sufficiently low packet loss rate, the use efficiency of the entire network will be reduced due to the limitation of the capacity of the default channel, such as the loss of notice packets etc. halfway without reaching the partner node and failure of communication. Problems arise. Therefore, when the capacity of the default channel is insufficient, a notice packet is preferentially transmitted for an instantaneous case, and normal data is handled by buffering or discarding. On the other hand, if there is an average shortage, a multi-default channel configuration in which a default channel is set for each ring is effective. FIG. 9 shows the configuration of the node in this case. Optical transceivers 108 and 109 for default channel
The WDM couplers 105-1 to 105-8 and the space switches 124-1 and 2 controlled by the control unit 111 enable the input and output of optical signals to and from the default channels of all rings. However, the optical transceiver 108 for the default channel is still used.
And 109 have only one at a time. The node examines the communication order while checking the load of the ring via the default channel in each ring. Immediately after that, the node that has switched the ring may broadcast a packet on the new ring that informs the user that the connection has been made and that informs information such as the communication status of the previous ring. As a result, almost the latest information of another ring can be obtained even in a certain ring, and this information is useful for scheduling such as switching between rings of another node.

By distributing the default channel to each ring, the default channel mentioned in the first embodiment and the bypass channel of 1.3 μm / 1.55 μm wavelength multiplexing can be used in the same manner. it can.

In the multi-default channel configuration,
When the node has a plurality of optical transceivers for the default channel, the optical transceiver of the node is somewhat expensive, but the communication efficiency can be further improved. This is because the control unit 111 can obtain, in addition to the ring currently communicating, the ring scheduled to be switched next and the default channel information of the next ring. Even if there are only a few such nodes in the network, that node can flow the latest other ring information and can be used as a reference for other nodes to schedule ring switching. Also, a node having a plurality of bypass channel optical transceivers 108 and 109 needs to have a plurality of default channel optical transceivers 106 and 107 in order to perform control for each optical transceiver. At this time, information exchange between the rings can be simultaneously assisted.

With the configuration shown in FIG. 10, it is more preferable that all the nodes include the default channel optical transceivers by the number of multi-default channels. The default channel optical transceivers 108-1 to 4 and 109-1 to 4 for each ring are W
It is connected to one of the optical rings via DM couplers 105-1 to 10-8. Each of the transmitting units 108-1 to 108-4 includes an ML unit 12 that performs a multilink connection control described later for each destination node.
A packet is input via 5-1 and 2. In addition, after the merger 118 for the normal default channel, there is also provided a sorter 127 for sorting packets for each destination with the above-mentioned ML unit. The output of each receiving unit is the ML unit 125-3,4
Are combined by the merger 126 via the, and passed to the selector 119.

With respect to the third problem that arises in this case, the fifth aspect of the present invention makes it possible to realize simple and efficient use of the default channel. First, by providing a function of branching communication between a plurality of default channels, and having a function of merging communication from a plurality of default channels, a plurality of default channels for each node can be connected to one link configured by multi-connection. It can be regarded as. Therefore, the control unit of the node does not need to distinguish which information is to be sent to which default channel connected to the other party. Each node selects a low-load default channel and sends out notice packets and information, distributing the load between each default channel, keeping the communication loss rate low enough, and reducing the efficiency of the entire network due to insufficient performance of the default channel. Is less likely to occur. Then, by applying an order control protocol such as a multilink protocol to the branching / merging function, the order of the communication reservation is not reversed, and highly reliable processing is realized. Further, even if a certain default channel becomes unavailable due to a failure, a highly reliable channel free from the failure can be realized by utilizing the remaining channels.

In the first and second embodiments, an example using "tell-and-go" as the communication setting method has been described. In this method, the procedure is simple, and each node can perform scheduling independently and autonomously. Communication settings can be adjusted and changed according to the status of each node. It also has the advantage of being able to cope freely with geographical and geographical deviations, and being very resistant to partial failures from being a distributed system. Moreover, the throughput is higher than that of the ALOHA method that sends data freely. However, when the load on the network increases, communication collisions occur frequently and the network cannot be used efficiently. Therefore, it is necessary to design a network with a sufficient margin in terms of communication capacity, and there is also a problem that costs are high.

On the other hand, the present invention does not rely on the communication setting method. For example, in order to more efficiently use the original communication capacity of the network, a pre-assignment type communication setting method in which the schedules of all nodes are determined in advance is effective, but the present invention is also effective in this case. . In the case of using this method, there is a scheduler that performs scheduling of the entire network, and the scheduler collects communication requests of the respective nodes, and performs scheduling of all nodes based on the collected requests. Even in this case, by adopting the configuration according to the first or second and third aspects of the present invention, the effect can still be obtained as described below.

First, the scheduler operates at the node (or
Optical transceiver for each bypass. The same applies to the following). Then, the group is assigned to one of the rings. In each ring, communication from node A in one group to node B in a different group. And scheduling of communication between nodes belonging to different groups. When the scheduling of communication between groups is completed in each ring, the next combination of groups is set in each ring, and communication between groups in that ring is scheduled. This is done sequentially, but when there is no opponent group,
Scheduling of communication within the group is also performed. At this time, the change in the combination of the groups is ring switching for each node, and the change in the partner node during inter-group communication is the wavelength switching. Therefore, here also
Type switching is distinguished, and scheduling is performed so as to reduce the frequency of ring switching (= space switching). As a result, network utilization efficiency is improved.

Also, by performing the two-stage scheduling in this manner, the scheduler only needs to schedule a combination of a group number that is much smaller than the number of nodes and a node number in the group that is always smaller than the number of nodes. Also, there is an advantage that the processing amount of is greatly reduced. Next, effects of the present invention in end-to-end communication performed over a plurality of networks will be described.
A third embodiment will be described with reference to FIG.

In FIG. 11, it is assumed that terminal A 205-1 transmits burst data to terminal B 205-2 via a plurality of optical networks 201, 202, and 203. Each network has
A plurality of nodes 102 are connected, and both terminal A and terminal B are connected to the network via one of those nodes. Intermediate networks I (201), II (202), I
II (203) is a network that effectively utilizes the wavelength and space dimensions, as shown in the first embodiment and the like. The terminal A connected to the network I sends a notice packet for communication to the terminal B to a default channel on the network I. The notice packet transferred on the default channel is transferred on the default channel of the network II via the gateway I-x206-1. Further, the notice packet is transferred on the default channel of the network III via the gateway II-x206-2, and reaches the terminal B. In the gateways Ix and II-x, the connection status of the network is fixed, and the channel indicated by the notice packet passes through the gateway and passes from the terminal A to the terminal B.
Exists as an optical path to If there is no particular problem in the process up to this point, the burst transmitted from terminal A after the notice packet is transmitted to terminal B.

FIG. 12 shows an example of a gateway node connecting networks I and II. The wavelength selector 123- connected to the rings # 1 and # 2 of each network
The optical signals selected by 2,3,4,5 are connected to the couplers 104-1 to 10-8 via the optical matrix switches 211-1 and 2 to one of the ring networks # 1 to # 4 of the other network. Inserted through. The optical matrix switch is a spatial optical switch that switches the optical path, an optical amplification function that amplifies an optical signal and shapes the waveform, an optical waveform regenerating function, and a device that converts the wavelength of an extracted optical signal to a different wavelength and inputs it to a new network. Wavelength conversion function. The light matrix is switched by the control unit via a control line from a control unit (not shown).

Generally, in the case of end-to-end, since the status of the network and the status of the terminal B cannot be grasped at the terminal A, the terminal A collects the basic information of the terminal B and the route to it in advance. Terminal A first sends a notice packet for establishing a communication schedule. This packet, which is slightly different from a normal notice packet, describes a source address and a destination address, a combination of an optical ring and a wavelength to be set for communication, an expected burst size, and its occurrence frequency. Based on this information, terminal B incorporates the communication from terminal A into the schedule.As a result, the terminal sets the combination of the optical ring and wavelength of the channel to be used for reception, the timing of waiting for a signal coming from terminal A there, and the period thereof. B notifies terminal A via the default channel. The intermediate nodes, particularly the gateway node and the terminal A, plan to incorporate the notified conditions into the scheduling. In the scheduling in the terminals A and B at this time, the control unit performs scheduling such that the frequency of ring switching is smaller than the frequency of wavelength switching, using the function characterized in claims 1-3 of the present invention. Do. As a result, the time spent for overall switching can be minimized, and efficient network utilization can be realized. According to the inventions according to the fourth and fifth aspects, transfer of control information and data can be easily and flexibly realized, and a highly reliable network is similarly realized. In addition, the terminal A sets the queue for the packet addressed to the terminal B in the destination channel queue for the bypass channel based on the notification, and sets new routing information also in the routing function. The settings made by the terminals A and B and the intermediate nodes are held in a soft state system so as not to waste communication resources. That is, the terminal A sends a burst to the terminal B periodically, or instead of the burst, sends a control packet requesting the setting retention to the terminal B, and transmits the burst to the terminal B.
It is necessary to continue to keep this setting for intermediate nodes. If these actions are not performed within a predetermined time, this setting is discarded at each terminal and node.

Next, based on this example, an example according to the seventh claim and its effect will be described.

As described above in connection with the fifth problem, when a burst is transferred across a plurality of networks as described above, the optical signal is transmitted from all the wavelengths of all the rings available to the terminal A to the terminal B. Cannot be sent out. This is because the number of rings and wavelengths connected by gateways on the way are limited. Therefore, the terminal A determines the communication condition from the ring and the wavelength transferred by the gateway. However, in the case of passing through a number of stages through the gateway, the choices of the communication conditions are reduced, and the case may not be the worst case. In such a case, if a wavelength converter is provided, the problem can be solved and communication can be performed efficiently. A wavelength converter transmits an optical signal that has entered at a certain wavelength instantaneously at another wavelength. At that time, information on the optical signal does not change. There are a type that converts electricity once and a type that converts light as it is, but none of them performs buffering of data inside the function. With this wavelength converter, the number of optical rings and wavelengths that can be connected to other networks can be independently set in each of a plurality of networks to which the gateway belongs. In addition, even if there is a collision of the used wavelengths, it can be avoided by shifting the wavelength of one of the optical signals.
As a result, each network can realize interconnection to another network with the optimal and minimum wavelength. Further, when used in combination with a multiplexer / demultiplexer, the ring can be switched by the wavelength converter, so that more efficient interconnection design and utilization can be achieved. In addition, by having both a wavelength converter and a space switch in the gateway,
Highly and efficiently use limited network connection capacity. The above-described gateway determines an appropriate output ring and wavelength in consideration of the state of the relay destination network when relaying the notice packet, rewrites the notice packet with the information, and transfers the information to the next node. At the final gateway II-X, the situation of the receiving terminal B is somewhat known, and the output ring and wavelength of the burst are determined based on the situation and the network situation, and the notice packet is transferred. Consider the same in the notification packet from terminal B.
adjust. Thereby, the flexibility of the communication setting is improved.

If the control unit of the gateway has such a high capability that such an advanced setting can be performed in real time, this control omits the above-mentioned pre-setting and the terminal A considers the situation of the other party. Even if the communication is started without any error, the success rate of the burst transfer is increased.

The space switch for switching the connection ring in the gateway is preferably different from the other nodes and has a sufficiently high switching speed. For example, a system such as a matrix switch using SOAG or a space switch using a combination of a multiplexer / demultiplexer and a wavelength variable light source is used. Because
The capacity of the communication path connecting the networks in the gateway is often limited, and it is necessary to use it efficiently. Moreover, in the gateway, since each terminal transmits the burst transmitted independently without temporarily storing, the switching of the wavelength and the connection between the rings occurs randomly,
Moreover, the schedule cannot be adjusted at the gateway. For this reason, it is better to reduce the guard time for switching even if it costs a little. Of course, the same can be said for the wavelength converter.

[0080]

As described above using some embodiments, in a large-capacity optical network in which a communication channel is dynamically set, the inefficient use of the communication capacity and the difficulty of controlling the communication capacity tend to occur. The problem is solved by the technology disclosed in the respective claims of the present invention, in which the setting of the switching order at the node, the selection of the data transfer channel, the use of the control information communication path, and the efficient channel connection at the gateway are performed. Thus, efficient and highly reliable data communication can be realized.

[Brief description of the drawings]

FIG. 1 is a first example of an optical node illustrating the present invention.

FIG. 2 is an example of a large-capacity network to which the present invention is applied.

FIG. 3 is an example of a multi-ring optical network showing the problems solved by the present invention.

FIG. 4 is a procedure example of point-to-point communication in a multi-optical ring network.

FIG. 5 shows an example of a scheduling algorithm for controlling a maximum buffer delay.

FIG. 6 is a frame configuration example of a notice packet.

FIG. 7 is an example of a frame configuration of a burst packet.

FIG. 8 is a second example of an optical node illustrating the present invention.

FIG. 9 is a first example of an optical node supporting a multi-default channel.

FIG. 10 shows a second example of an optical node supporting a multi-default channel.

FIG. 11 is a network configuration related to end-to-end communication.

FIG. 12 is a configuration example of a gateway node that connects networks.

[Explanation of symbols]

 102, 102-1, 102-2 ... Nodes 103-1 to 4 Optical ring 106 Burst optical receiver 107 Burst optical transmitter 108 Optical receiver for default channel 109 Optical transmitter for default channel 110 Space switch 111 Control unit 115 Routing unit 118 Junction unit 119 Selector 123 Wavelength selection unit 140 Forecast packet 150 Burst data packet 170 Forecast packet format 181 Burst packet format 124-1,2 Spatial optical switch 125-1 to 4 Multilink connection control unit 201,202,203 Network 205-1 to 3 Terminal 206 -1,2 gateway 211-1,2 optical matrix switch

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5K002 AA05 BA04 BA05 BA06 BA13 CA05 DA02 DA05 DA11 FA01 5K030 GA03 JA01 JL03 LA17 LE05 LE17 5K069 AA08 BA01 BA09 CB09 DB01 DB33 EA25 EA26 FA26 9A001 CC03

Claims (5)

[Claims] 1. An optical network for performing communication via a channel designated by space or wavelength or a combination of both, wherein a node belonging to the optical network and performing communication includes:
One or more optical transmission / reception means, two or more independent space / wavelength switching means for determining the space or wavelength used by the optical transceiver, and a series of communications performed by the node while changing the partner node with time. An optical network, comprising: a control unit that plans a communication schedule so as to preferentially use a unit having a short switching time among the two or more switching units. 2. The switching means is one space switching means and one wavelength switching means, the means having a short switching time is the wavelength switching means, and the control means frequently controls the wavelength switching means. 2. The optical network according to claim 1, wherein the communication network is used to switch a communication partner node. 3. The method according to claim 1, wherein the control means frequently uses the priority switching means and plans to switch the communication partner node so as to satisfy a standard for achieving a specific communication quality. The optical network according to claim 1. 4. An optical network for communicating via a channel designated by space or wavelength or a combination of both, wherein said optical network is a dedicated control information communication path for exchanging control information between nodes. A node belonging to the optical network and performing communication,
Communication schedule for performing one or more main optical transmission / reception means, space or wavelength switching means for determining a space or wavelength used by the main optical transceiver, and a series of communication performed by the node while changing the partner node with time. Means for exchanging control information between nodes via the control information communication path.
An optical network comprising: at least one optical transmitter / receiver; and means for joining / separating the control information and the data for transferring network user data via the control information communication path. 5. An optical network for performing communication via a channel designated by space or wavelength or a combination of both, wherein said optical network is a dedicated control information communication path for exchanging control information between nodes. And a node belonging to the optical network and performing communication is a main one.
At least one optical transmitting / receiving means, a space or wavelength switching means for determining the space or wavelength used by the main optical transceiver, and a communication schedule for performing a series of communication performed by the node while changing the partner node with time. Control means, a plurality of optical transceivers for the control means to exchange control information between nodes via the control information communication path, and the plurality of control information communication paths are regarded as one communication path. A function of transmitting data onto the communication path so as to average the communication load of each communication path, and a function of combining signals received from the plurality of control information communication paths. Optical network.