JPH08237253A - ATM network virtual path switching method and apparatus - Google Patents

Abstract

(57) [Summary] [Objective] In an ATM network in which a virtual path (VP) is set to a dual use of a working path and a backup path, it is possible to achieve synchronous switching without interruption during switching between the working and backup VPs. A switching control OAM cell with a sequence number and a switching necessity identification code is generated, and a switching control OAM cell is generated in a transmitting node 1 and a duplicate VP (V
P10T, VP11T), and the VP switching system 2-1 of the receiving node 2 uses the identification code and the sequence number of the switching control OAM cell to check the sequence number between the two VPs. As a result, the delay difference and the loss of the switching control OAM cell are detected, and the VP with which the synchronization is established is switched. The switching of the working VP and the backup VP of a plurality of VPs is managed independently. [Effect] It is possible to independently switch each VP by realizing non-instantaneous switching between VPs and managing the address of the cell stored in the synchronization establishment buffer memory for each VP even for a plurality of VPs.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of controlling switching of a virtual path (hereinafter abbreviated as VP) of an asynchronous transfer mode (hereinafter abbreviated as ATM) network and a switching device used therefor, more specifically, between nodes of an ATM network. Two VPs are set up, one of them is used as an active side and the other is used as a spare, and an active V is used by using a switching control management cell (hereinafter, abbreviated as an OAM cell for switching control).
The present invention relates to a method and apparatus for switching from P to backup VP.

[0002]

2. Description of the Related Art A method is known in which a VP is set dually between nodes of an ATM network, one of them is used as a working and the other is used as a spare, and a switching control OAM cell is used to switch between a working VP and a spare VP. For example, B-ISDN technology, Ohmsha, New Japan ITU Association, p. 185), but the specific technology has not been realized yet. In addition, V between nodes of the ATM network
When P is set to double, one is used as the current and the other is used as the backup, and the VP that is being selected is forcibly switched to the other VP, as a method of switching without causing a momentary interruption (non-instantaneous interruption switching), , "VP using NE linkage control in ATM network
Non-instantaneous switching system "1993 IEICE Spring Conference
As shown in B-763, switching control OAM cells are inserted in the cell streams of both systems, the delay time difference between the working VP and the backup VP is mutually notified, and this is used on the receiving side to use the working VP and the backup VP. A method has been proposed in which synchronization is established between them and then switching is performed.

[0003]

When the VPs are dually set between the nodes of the ATM network, the same cell is transmitted to each same VP, and the working VP and the backup VP are switched by receiving only. , ATM received from old working VP before and after switching
It is necessary to guarantee continuity between the cell and the ATM cell received from the new working VP. That is, it is necessary to prevent dropping or duplication of ATM cells by switching and to prevent instantaneous interruption of information transmitted by the ATM cells. In order to guarantee the continuity of ATM cells, the switching must be executed after absorbing the transmission delay difference existing between the working VP and the backup VP. Furthermore, the VP in which each node is involved is not limited to a single VP, and is usually connected to a plurality of VPs that are multiplexed. In a network with such VP settings,
When a specific VP requires switching and another VP does not require switching, it is necessary to individually control switching of the multiple VPs.

Further, when the set VP is used for bidirectional communication, the V selected by both nodes at the VP end is set.
P must match. That is, the VP is set to be bidirectional and dual between the nodes A and B, and when one of the VPs set to be dual is set to the 0 system and the other is set to the 1 system, the node A is set to 0. If the system is being selected, the node B also needs to select the 0 system.

In the above-mentioned conventionally known technique, a switching OAM cell is used in order to detect the delay time difference between the working and protection VPs. That is, at the transmitting end, the switching OAM cells are sent to the working and protection VPs, and at the VP receiving end due to the time difference between the receiving switching OAM cells in both systems,
This is a method in which the delay time difference is detected and the VP cells of both systems are switched after establishing synchronization.

[0006] However, the above-mentioned conventional technique is not described for confirming the normality of the spare VP. That is, even when the backup VP is abnormal, when the device receives the switching command, the device may switch from the working VP to the backup VP. Also, switching OAM
In the method of detecting the cell delay time difference between both VPs using the cell, when the switching OAM cell is dropped, it becomes impossible for the receiving node to detect the delay time difference between the working VP and the backup VP, and the cell dropout occurs. However, the above-mentioned conventional technique does not mention the technique for detecting the loss of the switching OAM cell. Also, AT in the transmission line
It is generally known that fluctuation occurs in the transmission time of the M cell due to the switching process in the relay node, but the fluctuation is not mentioned in the above-mentioned conventional technique. Further, in the above conventional technique, the node at one end of the VP is
Neither the technique of confirming the VP system selected by the node at the other end nor the technique of confirming that the VP systems selected by the nodes at both ends of the VP match each other. That is, a technique for ensuring that nodes at both ends of the VP select the same system VP has not been studied. Furthermore, the above-mentioned conventional techniques only deal with the processing of a single VP. That is, a switching processing technique for a plurality of VPs multiplexed on the same line has not been studied.

A first object of the present invention is to provide a method and a switching device for synchronizing VPs which are set to be duplicated, and for switching a VP in a receiving node so that a cell does not drop downstream. Is to be realized. Second of the present invention
The purpose of is to realize a means for confirming the normality of the standby system at the time of executing the switching when the switching instruction is received. A third object of the present invention is to realize an apparatus and method for guaranteeing that, when a switching control OAM cell is lost, the loss or duplication of information cells does not occur due to the loss. A fourth object of the present invention is to guarantee that the cell does not drop even when fluctuation occurs in the propagation delay time of the cell on the transmission path. The fifth object of the present invention is to
It is to ensure that the nodes at both ends of the VP select the same system VP. A sixth object of the present invention is to realize a VP house price change by keeping the working VP and the backup VP in synchronization with each other when a plurality of connections are set on the same line.

[0008]

In order to achieve the above object, a virtual path switching control method for an ATM network according to the present invention uses a switching control OAM cell including a switching command code for identifying the presence or absence of a switching request and a sequence number. , The transmitting side node transmits the switching control OAM cells to both the working and standby sides in parallel and intermittently, and the receiving side node compares and collates the sequence numbers of the switching control OAM cells, so that both VPs are When the phase of the switching control OAM cell that arrives is detected and switching is instructed by the switching control OAM cell, switching between the working VP and the backup VP is performed in a form that compensates for the phase.

Further, in order to execute the switching only when the standby VP is normal, the switching processing is executed after confirming the normality of the OAM cell for switching control which arrives from the standby VP. In addition, when the receiving node detects an abnormality in the backup VP even during the switching process, it ignores the instruction of the OAM cell for switching control and continues to select the cell of the working VP. The receiving end node detects the loss of the switching control OAM cell by verifying the sequence number of the switching control OAM cell, and when the loss is detected, the switching is stopped. Further, in order to realize the switching considering the fluctuation, the switching processing is executed by using the switching algorithm corresponding to each of the existing fluctuation patterns. Further, the switching control OAM cell is provided with an area (currently selected system identifier) for displaying the number of the VP currently selected by the node at the VP end.
The node of one end of the switch displays the system number of the selected VP in the switching control OAM cell, and transmits the switching control OAM cell to both the working VP and the backup VP in parallel and intermittently, and the receiving side The node compares and collates the system number of the received OAM cell for switching control with the system number of the VP being selected by the receiving side node, so that nodes at both ends of the VP select the VP of the same system. Verify that you are doing.

In the present invention, as a method of establishing the synchronization of the received cells from the working VP and the backup VP, (1) establishment of synchronization at the time of switching, which establishes synchronization at the time when the switching instruction code instructs switching, and (2) ) Regardless of the contents of the switching command code, the number of cells received by the working VP and the backup VP is always counted, and only the number of arriving cells of the VP having the smaller count value has arrived at both VPs. As for, there is a constant synchronization establishment for establishing a constant synchronization.

In order to carry out the above virtual path switching control method, the virtual path switching apparatus of the present invention comprises an extracting means for extracting the switching command code and sequence number of the above switching control OAM cell from the working VP and the backup VP. , The working and spare V depending on the extracted sequence number
Means for synchronizing the phase of P, means for storing a switching algorithm corresponding to each of the delay patterns (cell arrival phases) of the working and protection transmission cells, and the working and protection when the extracted switching command code is a switching request. It has a switching unit that performs a switching process according to a switching algorithm corresponding to the arrival phase of the cell.

Further, when a plurality of VPs are set, a means for managing the address of the buffer memory for each VP by using the switching instruction code and the sequence number when writing the arrived cell into the buffer memory once. Set up.

[0013]

In the present invention, the switching control OAM cell has the switching command code and the sequence number for indicating the presence / absence of the switching request, and is transmitted intermittently. Therefore, the receiving side node can distinguish between the working VP and the backup VP. By checking the sequence number, it is possible to easily detect the delay difference between the switching control OAM cells of the working VP and the backup VP and the presence or absence of the switching control OAM cell being dropped. Therefore, it is possible to accurately perform the synchronization at the time of switching the working VP and the backup VP. Therefore, it is possible to prevent the information cells from being dropped or duplicated when switching.

Further, in the present invention, the switching control OAM cell is transmitted to both the working VP and the backup VP, but the switching command displayed in the switching control OAM cell arriving at the backup VP on the receiving side is abnormal. If there is, switching is not executed. With this, after confirming the normality of the backup VP,
Switching from the working VP to the backup VP can be realized.

Further, in order to deal with the fluctuations of the working VP and the backup VP, a delay fluctuation pattern is defined. That is, (1) when the switching control OAM cells arrive in the order of the working VP, the backup VP, the working VP, and the backup VP, (2) the working V
O for switching control in the order of P, spare VP, spare VP, and working VP.
When the AM cell arrives, (3) the backup VP, the working VP,
When the switching control OAM cell arrives in the order of the spare VP and the working VP, (4) 4 in the case where the switching control OAM cell arrives in the order of the spare VP, the working VP, the working VP, and the spare VP
Define the pattern. Even if delay fluctuations occur by specifying the switching algorithm corresponding to each,
Switching can be performed without dropping cells or duplication.

Further, the switching control OMA cell is provided with an identifier for displaying the VP number selected by the transmitting node in addition to the switching command code and the sequence number, and the switching control OMA cell is switched. By transmitting the identifier intermittently and transmitting the identifier to the other end of the VP even at times other than the time, when the node at one end of the VP is playing the band of the selected VP, the node at the other end of the VP selects the The nodes on both ends of the VP can mutually confirm the coincidence of the numbers.

Further, even when the receiving node receives information cells from different transmitting nodes from a plurality of VPs, the sequence number of the switching control OAM cell and the information on the presence / absence of the switching request are used to perform the above. Switching can be performed independently by managing the address of the buffer memory for each VP.

Further, the switching control OAM at the transmitting side node
A sequence number is given to the cell, the sequence number is stored in the OAM cell for switching control received by the receiving side node, the sequence number is given to the OAM cell for receiving switching control in the receiving side node, and stored. By comparing and collating with the sequence number of the newly arrived switching control OAM cell, the switching control OAM cell loss can be easily detected in any of the switching synchronization establishment method and the constant synchronization establishment method. Can take measures.

[0019]

1 is a block diagram of a VP switching system of an ATM network in which a VP switching control method according to the present invention is implemented. The figure shows a dual VP (VP10R, VP10T and VP11T, VP1) between a node 1 and a node 2 in an ATM network.
1R), a double VP between node 5 and node 2 (VP
20R, VP20T and VP21T, VP21R) are formed. Nodes 3, 4 and 6 are relay nodes. In the above VP symbol, the last R and T indicate reception and transmission, respectively. A terminal and another network are connected to the nodes 1 and 2, respectively. Double VP
One is used as the current one and the other is used as a spare. Between the nodes 3 and 2, a plurality of Vs (two are shown for simplicity).
The state in which P (VP10 and VP20) is formed is shown.

A switching control O is provided to each of the nodes 1 and 5.
AM cell generation / insertion circuits 1-1 and 5-1 are provided. The generation of the switching control OAM cells occurs continuously, that is, intermittently at intervals of the switching control OAM cells of the format described in FIG. The transmission control node transmits the switching control OAM cell to both the working and protection VPs in parallel. In the VP switching device 2-1 of the receiving side node 2, the working and protection OAM cells for switching control are detected and compared, and the switching command code switching request in the switching control OAM cells, the switching control OAM cell Phase absorption is started using the middle serial number, and the active and standby VPs are switched at the boundary of the switching control OAM cell.

FIG. 2 shows the format of a switching control OAM cell used in an embodiment of the present invention. O for switching control
The AM cell includes a header c-1 including a VP identifier, an OAM cell type c-2, a switching instruction code c-3, a failure detection system identifier c-4, a currently selected system identifier c-5, and a sequence number c.
-6, unused area c-7, CRC-8 area. O
Areas c-1, c-2, c-7, c-8 in the AM cell configuration
Is the same as that conventionally known. Here, as the switching instruction code c-3, there is no switching request (No Request: N
R) and switching request (Forced Switch: FS) are defined.

FIG. 3 is a block diagram showing the configuration of a VP switching device of a receiving node used in the first embodiment of the VP switching method according to the present invention. The switching control method used in the present embodiment is a switching synchronization establishment method that establishes synchronization when a switching request occurs. Further, when a plurality of VPs are set, the VPs can be independently switched for each VP.

Dual input highways 23-1 and 23-
2 connected to VP filters 8-1 and 8-2, respectively
The switch control OAM cell is extracted and added to the buffer memory control unit 9 and the switch control unit 12. In addition, the VP filter 8
-1 and 8-2 are sent to the switching control buffer memory 7 for each VP according to an instruction from the switching control unit 12, or
It is selected whether to send to the output lines 24-1 and 24-2.
The buffer memory control unit 9 uses the VP filters 8-1 and 8-
The writing and reading of the buffer memory 7 are controlled by the VP identifier from 2 and an instruction from the switching control unit 12 (details of the buffer memory control unit 9 will be described with reference to FIG. 4). The output lines 24-1 and 24-2 are provided with empty cell detection units 16-1 and 16-2, respectively, and information of empty cells is added to the switching control unit 12. The selector 13 connects one of the cells of the two systems via the output lines 24-1 and 24-2 and the cell of the output of the buffer memory 7 to the output highway 25. The cell selected by the selector 13 is output to the output highway 25 at a predetermined transmission rate via the memory 17. The switching control unit 12 is shown in FIG.
The cells arriving at the working VP and the backup VP are managed for each VP according to the algorithm described with reference to FIG. It is determined whether or not the belonging cell is stored in the switching control buffer memory 7, and this is determined by the buffer memory control unit 9 and the selector 1.
Add to 3.

FIG. 4 is a block diagram showing the configurations of the buffer memory 7 and the buffer memory controller 9 shown in FIG. This configuration uses the address chain method because the buffer memory 7 is managed for each VP. A set of two types of registers corresponding to a plurality of VPs to be managed (write registers 10-1 to 10-4 and read registers 11-1 to 11-
4) The same number of VPs that manage
The data output of the IFO memory 15 is connected to the input terminals of the respective write registers 10-1 to 10-4 and the data input terminal of the buffer memory 7, and the output terminals of the plurality of write registers 10-1 to 10-4 are buffers. Memory 7
Connect to the write address pin of. Write register 10
-1 to 10-4 are a plurality of VP # 1 to VP # 4, respectively.
Corresponding to. Also, the read registers 11-1 to 11-4
Is connected to the read address terminal of the buffer memory 7. The read registers 11-1 to 11-4 correspond to VP # 1 to VP # 4, respectively. Here, for simplicity, the number of VPs to be managed is four. The empty address FIFO memory 15 stores the address value of the unused area of the buffer memory 7 detected by the empty cell detection unit 16. Further, an access control unit 14 that controls writing and reading of these is provided.

On the output side of the buffer memory 7, the data output terminal of the buffer memory 7 is connected to the input terminals of the respective read registers 11-1 to 11-4, and the plurality of read registers 11-1 to 11-4 are connected. The output terminal is
The output of the access control unit 14 that generates the read timing for each VP is connected to the read address terminal of the buffer memory 7, and the read registers 11-1 to 11-
The output of 4 is also connected to the data input terminal of the empty address FIFO memory 15.

In the buffer memory 7, a cell and an address (next address) for storing a cell belonging to the same VP as the cell and arriving next to the cell are stored at the same address. With this circuit, arriving cells can be stored in the buffer memory 7 and cells can be sequentially read from the buffer memory 7 for each VP. Therefore, according to the algorithm described below, the switching control unit 12 stores the cell belonging to the VP that requires the delay insertion in the buffer memory 7, and after the switching control OAM cell arrives in another system, The cell belonging to that VP is read. The reading is performed when the empty cell detection units 16-1 and 16-2 detect empty cells (dummy data) in the highways 24-1 and 24-2, and the output highway 25 is replaced with the empty cells. Output.

When storing a cell in the buffer memory 7, the address is read from the write address register corresponding to the VP to which the cell belongs, and this is read.
It is given to the write address terminal of. At the same time, a vacant address is read from the vacant address FIFO memory 15 and written in the buffer memory 7 together with the above cell.

When reading a cell from the buffer memory 7, the address is read from the read register corresponding to the desired VP, this is given to the read address terminal of the buffer memory 7, and is stored in the empty address FIFO memory 15. By this operation, the cell and the next address are read from the buffer memory 7. The next address is stored in the read address register 11. As a method of determining the reading order of a plurality of VPs, the number of times of reading may be determined in proportion to the occupied band of the VPs.

FIGS. 5 to 12 are diagrams for explaining the algorithm of the switching method by dividing the relation between the arrival times of the switching control OAM cell of the working VP and the switching control OAM cell of the backup VP. To simplify the description, the case of a working VP and a backup VP between two nodes will be described.
Further, as a precondition, the insertion interval of the switching control OAM cells is set to be larger than the delay difference between the two systems. Further, the capacity of the buffer memory 7 for delay insertion is larger than the number of cells corresponding to the delay difference between the working VP and the backup VP, and the switching control OA
It is smaller than the number of cells corresponding to the insertion interval of M cells.

FIG. 5 shows that the receiving node has a working VP and a backup V
A case where P cells arrive at the same time (there is no delay difference) is shown. In the figure, the area of NR or FS is the switching control OA.
It is an M cell, and the display of NR and FS indicates that the contents of the switching command code c-3 are no switching and switching request, respectively. In addition, n-1, n, n + in the lower stage of NR or FS
1 and the like are sequence numbers c-6 of the switching control OAM cell. Further, the area to be discarded is an empty or information cell (the same applies to FIGS. 6 to 12 below). In the case of FIG. 5, after comparing the switching command FS of the working VP and the backup VP and the sequence number n (), and confirming their match,
Perform switching immediately (). This state is switching mode 1
And

FIG. 6 shows a case where the cell of the backup VP arrives earlier than the cell of the working VP. The cells after the switching control OAM cell (FS, n) arriving from the spare VP are stored in the buffer memory 7 (,), and the working V having the same sequence number n as the switching control OAM cell of the spare VP is stored.
When the OAM cell for P switching control arrives (), the working VP is disconnected, switched to the backup VP (), and output from the buffer memory 7 is started (). Reading from the buffer memory 7 is performed at a speed higher than the nominal speed of the VP. As a result, the buffer memory 7 is separated when the buffer memory 7 becomes empty. This state is referred to as switching mode 2.

FIG. 7 shows a case where the working VP cell arrives earlier than the backup VP cell. As shown in FIG. 7A, the sequence number (n) of the switching control OAM cell arriving from the working VP is temporarily stored in the register (), and then the backup VP is switched with the same sequence number (n). After receiving the control OAM cell (,), then the working V
OAM cell for P switching control (sequence number: n + 1)
When () arrives (), the cell flow of the working VP is connected to the buffer memory 7 (), and when the switching control OAM cell (sequence number: n + 1) arrives at the spare VP, the spare V
Switch to P (). This state is called switching mode 3-1.

Further, as shown in FIG. 7B, the switching control unit 12 temporarily stores the sequence number (n) of the switching control OAM cell arriving from the working VP in the register (),
After receiving the switching control OAM cell having the same sequence number (n) from the backup VP (,), then the backup VP
When the switch control OAM cell (sequence number: n + 1) arrives (), the cell flow of the spare VP is connected to the buffer memory 7 (), and the switch control OAM cell (sequence number: n + 1) becomes the active system. When it arrives (), it switches to the backup VP and starts output from the buffer memory 7 (). This state is referred to as switching mode 3-2.

Next, in each of the above-mentioned switching modes, the processing when the switching control OAM cell is lost will be described below.
FIG. 8A shows processing in the switching mode 1 when the switching control OAM cell of the working VP is lost. Reserve V
Since only the P switching OAM cell (sequence number: n) arrives (), the switching control unit 12 determines that the current state is the switching mode 2, and connects the backup VP to the buffer memory 7 (). If a buffer overflow occurs before the switching control OAM cell (sequence number: n + 1) of the working VP arrives (), it is regarded as a switching control OAM cell loss, and the switching operation is canceled (), and the buffer is released. Clear the memory 7. Alternatively, even when the buffer overflow does not occur, the loss of the switching control OAM cell can be easily detected by comparing the sequence numbers of the switching control OAM cells that arrive next.

FIG. 8B shows the processing in the switching mode 1 when the switching control OAM cell (sequence number n) of the spare VP is lost. In this case, the switching control unit 1
2 determines that the current state is the switching mode 3, and the sequence number (n) of the switching control OAM cell arriving from the working VP.
Is temporarily stored in the register (). Then, by comparing with the sequence number of the switching control OAM cell of the spare VP that arrives next, it is confirmed that the sequence numbers do not match (), the loss of the switching control OAM cell is detected, and the mode is switched to the switching mode 2. To ().

Alternatively, as shown in FIG. 8 (c), although the switching mode 1 is originally used, the switching control OAM cell with the sequence number n from the backup VP is lost, so the sequence number n of the working VP is lost. When the OAM cell for switching control is received, the sequence number n is set in the register (), and the sequence number n + 1 of the working VP is received,
The loss of the switching control OAM cell having the sequence number n from the backup VP is known, and the process immediately shifts to the switching mode 3 (). When the sequence number n + 1 of the working VP is received, the switching control O of the sequence number n from the spare VP is received.
The reason why the AM cell loss can be determined is that the insertion interval of the switching control OAM cells is set to be larger than the delay difference between the two VPs, as described above.

FIG. 9 shows the processing when the switching control OAM cell in the switching mode 2 is lost. FIG. 9A shows a case where the switching control OAM cell of the working VP is lost, and the switching control OAM cell arrives at the standby system, and the switching control unit 12
Determines that the switching mode is 2 and starts the switching operation (), connects to the buffer memory (), and starts storing the cell of the spare VP. Next, since the OAM cell for switching control of the working VP is lost, a buffer overflow occurs (). The switching control unit 12 detects a buffer overflow and suspends the switching operation (). Or even if the buffer overflow is not detected,
Sequence number n of the switching control OAM cell that arrives next
By comparing +1, the loss of the switching control OAM cell can be easily detected.

FIG. 9B shows a case in which a loss occurs in the switching control OAM cell (sequence number: n) of the first switching request of the backup VP, and the switching control unit 12 receives the next working current. The sequence number n of the switching control OAM cell of the first switching request is stored in the register (), and the switching mode 3 is determined. Next, the switching operation is started in the switching control OAM cell (sequence number n + 1) of the switching request arriving at the backup VP (), and the switching control OAM cell sequence number n
By comparing +1 with the sequence number n of the working arriving switching control OAM cell, the loss of the switching control OAM cell of the spare VP is detected, and the spare cell is connected to the buffer cell 7 (). , The switching mode 2 is performed.

FIG. 9C shows the same case as FIG. 9B, in which the switching control unit 12 stores the sequence number n of the switching control OAM cell of the first active switching request that arrives next in the register ( ), From the above-mentioned prerequisites, when the switching control cell (sequence number: n + 1) of the working VP is received, the loss of the switching control OAM cell (sequence number: n) of the spare VP is detected, and the sequence number n of the register is detected. Is updated to N + 1 (), and then the switching mode 3 is performed.

FIG. 10 shows the processing when the switching control OAM cell in the switching mode 3 is lost. Figure 10
(A) shows the case where the switching control OAM cell (sequence number: n) of the working VP is lost. In this case, the OAM cell for switching control of the backup VP (sequence number: n) arrives next, so the switching control unit 12 determines that the switching mode is 2, and starts the switching operation (), and buffers the cell of the backup VP. Store in memory 7 (). The cells are accumulated until the next switching control OAM cell arrives at the working VP, and when a buffer overflow occurs, the switching control OAM of the working VP is performed.
It is judged that the cell is lost, the buffer memory is cleared, and the switching operation is stopped (). Even when the buffer overflow does not occur, the loss of the switching control OAM cell can be easily detected by the sequence number n + 1 of the switching control OAM cell that arrives next.

FIG. 10B shows an O for switching control of the spare VP.
The case where the AM cell (sequence number: n) is lost is shown. In this case, when the switching control OAM cell (sequence number: n) arrives at the working VP, the switching control unit 12 determines that the switching mode is 3, sets the sequence number: n in the register, and starts the switching operation (). The switching control cell of the next spare VP is lost, and then the switching control OAM cell (sequence number: n + 1) of the working VP arrives. In this case, the switching control OAM cell of the working VP (sequence number: n +
When 1) is received, it is determined that the switching control OAM cell has been lost (), the sequence number: n is set in the register, and the switching mode 3 processing is performed.

FIG. 10C shows a case where the OAM cell (sequence number: n + 1) for switching control of the spare VP has arrived in the same case as FIG. 10B. In this case, the switching control OAM cell of the working VP (sequence number: n)
Sequence number: n is set in the register (), and when the OAM cell for switching control of the spare VP (sequence number: n + 1) arrives next, the OAM for switching control of the working VP that arrived first Cell (sequence number: n)
And the sequence number are compared (), a mismatch is detected, it is determined that the OAM cell loss for switching control has occurred, and the mode is changed to the switching mode 3 ().

FIG. 11 shows the case where the switching control OAM cell is lost in the switching mode 3-1. FIG. 11A shows
In the switching mode 3-1, the case where the switching control OAM cell (sequence number: n + 1) of the working VP that should arrive next is lost. In this case, when the sequence number n arrives at the working VP, the sequence number n is set in the register (), and the sequence number is compared when the OAM cell for switching control of the spare VP (sequence number: n) arrives (). , So that the switching operation of the switching mode is started (). Next, OA for switching control of the working VP
M cell (sequence number: n + 1) does not arrive, so backup V
OAM cell for P switching control (sequence number: n + 1)
Has arrived, the switching control unit 12 determines that the switching mode is 3-2 (), and connects the highway output of the spare VP to the buffer memory 7 (). When the cell overflows from the buffer memory 7 (), the loss of the switching control OAM cell of the working VP is detected. Alternatively, even when the buffer overflow does not occur, the switching control OAM cell of the working VP (sequence number:
n + 1) loss can be detected. O for switching control of working VP
After the loss of the AM cell (sequence number: n + 1) is detected, the previous switching process is stopped (), and the switching control O of the working VP is stopped.
The switching process of the switching mode 3-1 is newly restarted from the AM cell (sequence number: n + 2).

FIG. 11B shows a case where the OAM cell for switching control of the backup VP (sequence number: n + 1) is lost in the same case as in FIG. 11A. The processing (,,) up to the OAM cell for switching control of the spare VP (sequence number: n) is the same as that in the case of FIG. 11B. After that, the sequence numbers are compared when the active switching control OAM cell (sequence number: n + 1) is received (), and the active VP cell is connected to the buffer memory 7 (). When the backup VP switching control OAM cell that should arrive next is lost, the active cell stored in the buffer memory 7 may be lost. Here, a threshold is set for the capacity of the buffer memory 7, and when the cell storage amount exceeds this (), it is determined that the switching control OAM cell is lost, and the output of the buffer memory 7 is connected to the output highway. (), The cell of the working VP stored in the buffer memory 7 is read until the buffer memory 7 becomes empty. Further, even when the threshold value is not detected, the switching control OAM cell that arrives next detects the switching control OAM cell loss, and the above method is used to recover.

FIG. 12 shows a case where the switching control OAM cell in the switching mode 3-2 is lost. Figure 12 (a)
Is the backup V that should arrive next in the switching mode 3-2.
OAM cell for P switching control (sequence number: n + 1)
Shows the case of loss. In this case, the switching control OAM cell of the next working VP (sequence number: n +
According to 1), the switching control unit 12 determines that the switching mode is 3-1 (), and connects the cell of the working VP to the buffer memory 7 (). Then, the storage amount of the cell is equal to the buffer memory 7
When the threshold value set in advance is exceeded (), the spare control VP switching control OAM cell (sequence number: n +
1) Loss is detected, the previous switching operation is stopped, and the working VP is
Connected to the output highway 25, and the cells are read from the buffer memory 7 until the buffer memory 7 becomes empty ().
Alternatively, even when the buffer overflow does not occur, the loss of the switching control OAM cell can be easily detected by the sequence number (n + 2) of the switching control OAM cell that arrives next.

FIG. 12B shows the working VP that should arrive next.
2 shows a case where the switching control OAM cell (sequence number: n + 1) is lost. OA for switching control of spare VP
When M cells (sequence number: n + 1) are received, the sequence numbers are compared (), and the cells of the spare VP are connected to the buffer memory 7 (). When the switching control OAM cell of the working VP that should arrive next is lost, the buffer overflows (), so it is determined that the switching control OAM cell has been lost, and the switching operation is stopped (). Even when the buffer overflow is not detected, the switching control OAM cell loss of the working VP is detected by the next working switching control OAM cell (sequence number: n + 2), and the method is restored by the above method.

FIG. 13 is a block diagram showing the configuration of a receiving side node switching device used in another embodiment of the switching method according to the present invention. The switching control method of the present embodiment is a method of always synchronizing the working VP and the backup VP. Also in this case, the transmitting node intermittently transmits the switching control OAM cell shown in FIG. 2 by the working VP and the backup VP.

At the receiving node, cells that have arrived independently on the highways 23-1 and 23-1 of the working VP and the backup VP are buffer memories 7-1 and 7-2 via the VP filters 8-1 and 8-2, respectively. Remember. From the time when the switching control OAM cell of each VP arrives, the number of arriving cells is counted for each VP. Then, the number of received cells in the working VP and the number of received cells in the backup VP are compared, and the number of received cells of the smaller VP is made readable for both VPs. Then, the readable cells are read from the buffer memories 7-1 and 7-2 at the same time for both systems, and the selector 1
Control 3 and select one. By doing so, it is possible to switch after establishing synchronization at all times. The number of memory cells is counted by the counter 18-1 ... 18-for each VP to be managed.
4 and 19-1 ... 19-4 may be provided.

Further, since the sequence number c-6 is assigned to the switching control OAM cell as described with reference to FIG. 2, the switching control OAM cell loss can be easily detected. Further, by periodically transmitting the switching control OAM cells, the cell loss can be easily detected by independently counting the number of cells received during the switching control OAM cell reception at the reception point by the working VP and the backup VP. Buffer memory 7-
The control methods 1 and 7-2 may be performed in the same manner as in the first embodiment.

[0050]

According to the present invention, an OAM cell for switching control to which an identification code for VP switching necessity and a sequence number is added is intermittently transmitted, so that an information cell at the time of switching can be obtained while synchronizing active and standby switching. It is possible to prevent a momentary interruption due to the dropout.

[Brief description of drawings]

FIG. 1 is a schematic diagram of a VP switching control method according to the present invention.
Block diagram of VP switching system for TM network

FIG. 2 is a switching control OA used in an embodiment of the present invention.
Format of M cell

FIG. 3 is a block diagram showing the configuration of a VP switching device of a receiving node used in the first embodiment of the VP switching method according to the present invention.

FIG. 4 is a block diagram showing the configurations of a buffer memory 7 and a buffer memory control unit 9 shown in FIG.

FIG. 5 is an explanatory diagram of mode 1 of the switching algorithm in the embodiment of the VP switching method according to the present invention.

FIG. 6 is an explanatory diagram of mode 2 of the switching algorithm in the embodiment of the VP switching method according to the present invention.

FIG. 7 is an explanatory diagram of mode 3 of the switching algorithm in the embodiment of the VP switching method according to the present invention.

FIG. 8 is an explanatory diagram of a switching algorithm in the case of a switching control OAM cell loss in mode 1 of the switching algorithm.

FIG. 9 is an explanatory diagram of a switching algorithm when a switching control OAM cell is lost in mode 2 of the switching algorithm.

FIG. 10 is an explanatory diagram of a switching algorithm when a switching control OAM cell is lost in mode 3 of the switching algorithm.

FIG. 11 is an explanatory diagram of a switching algorithm when a switching control OAM cell is lost in mode 3 of the switching algorithm.

FIG. 12 is an explanatory diagram of a switching algorithm in the case of a switching control OAM cell loss in mode 3 of the switching algorithm.

FIG. 13 is a block diagram showing the configuration of a switching device used in another embodiment of the VP switching method according to the present invention.

[Explanation of symbols]

 1 to 6 ... Node 1-1 ... OAM generation / insertion circuit for switching control 2-1 ... VP switching device 5-1 ... OAM generation / insertion circuit for switching control 7, 7-1, 7-2 ... Buffer memory 8-1, 8-2 ... VP filter 9, 9-1, 9-2 ... Buffer memory control unit 10-1 to 10-4 ... Write address register 11-1 to 11-4 ... Read address register 12 ... Switching control unit 13 ... Selector 14 ... Access control unit 15 ... Free address FIFO memory 16-1, 16-2 ... Free cell detection unit 17 ... Memory 18-1 to 18-4 ... Counter 19-1 to 19-4 ... Counter 20-1 and 20- 2 ... Read control signals 21-1, 21-2 ... Write control signals 23-25 ... Highway

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Nobuhiro Ashi 216 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Hitachi, Ltd. Information & Communication Division (72) Inventor Hiromi Ueda 1-6 Uchiyuki-cho, Chiyoda-ku, Tokyo No. Japan Telegraph and Telephone Corporation (72) Inventor Hitoshi Uematsu 1-6, Uchisaiwaicho, Chiyoda-ku, Tokyo Japan Telegraph and Telephone Corporation (72) Inventor Hiroshi Ota 1-1-6, Uchisaiwaicho, Chiyoda-ku, Tokyo No. Japan Telegraph and Telephone Corporation

Claims (11)

[Claims] 1. A control command code and a sequence number for identifying the presence / absence of a virtual path switching request from the transmitting node by setting two virtual paths, a working path and a backup path, between a transmitting node and a receiving node in an ATM network. The switching control OAM cells that are transmitted to the virtual paths of the two systems are transmitted in parallel and intermittently, and are received by the receiving node.
The sequence number of the switching control OAM cell from the virtual path of the system is compared and collated, the phase of the switching control OAM cell is detected, and when the switching request is made, the working virtual path and the backup virtual path are switched in a form in which the phase is compensated. A virtual path switching method for an ATM network characterized by the following. 2. A control command code, a sequence number, and a sequence number for identifying the presence / absence of a virtual path switching request from the transmitting node by setting two virtual paths, a working path and a standby path, between a transmitting node and a receiving node in an ATM network. A switching control OAM cell having a currently selected system identifier indicating which of the two paths the virtual path is selected by the node is transmitted in parallel and intermittently to the two paths of the virtual path, and the receiving node O for switching control received
An ATM characterized by verifying the coincidence between the virtual path system selected by the receiving node and the system selected by the transmitting node by the currently selected system identifier of the AM cell.
Virtual path switching method of network. 3. The switching control O in the receiving node
When receiving an AM cell, the sequence number of the received switching control OAM cell for each virtual path is compared with the sequence number of the newly received switching control OAM cell of the same virtual path to determine the switching control OAM cell. The virtual path switching method for an ATM network according to claim 1, wherein loss is detected. 4. The switching control O in the receiving node.
2. When receiving an AM cell, the loss of the switching control OAM cell is detected by comparing the sequence numbers of the two-system virtual paths of the received switching control OAM cell. Virtual path switching method. 5. In the receiving node, when switching an old working virtual path to a new working virtual path, cells arriving from the old working virtual path are accumulated until the switching is completed, and when the switching is successful, the accumulation is carried out. 2. The virtual path of the ATM network according to claim 1, wherein the cell arriving from the old working virtual path is discarded, and when the switching fails, the accumulated old working virtual path is selected again and transmitted downstream. Switching method. 6. The receiving node stores cells for working and protection virtual paths in a buffer memory, and counts cells for working and protection virtual paths from the time when the switching control OAM cells having the same sequence number are received. Read out from the buffer memory for the same number of cells as the count value is smaller, and the same number of working and spare cells are read and transmitted each time the switching control OAM cell is received. Claim 1 characterized by the above.
The virtual path switching method of the ATM network described. 7. When the switching control OAM cell sent to the backup virtual path at the receiving node arrives earlier than the switching control OAM cell sent to the working virtual path at the receiving node, the switching control OAM cell is sent to the backup virtual path. The switching control OA which has started the switching operation according to the switching control OAM cell switching command and has been sent to the working virtual path.
If the M cell arrives at the receiving side early, the sequence number of the switching control OAM cell sent to the working virtual path is stored, and the switching control OAM cell arriving from the backup virtual path is waited for. The virtual path switching method according to claim 1, wherein the switching operation is started after confirming that the sequence numbers match. 8. A virtual path is provided between two nodes of an ATM network.
System setting, the transmitting node sends switching control OAM cells to the two virtual paths in parallel, and the receiving node receives the switching control OAM cells so that one of the two system virtual paths is in use and the other is in use. An ATM system for switching a virtual path as a backup, wherein the transmitting node transmits a switching control OAM cell having a control command code and a sequence number for identifying the presence / absence of a switching request in parallel and intermittently to the two virtual paths. A virtual path switching system for ATM networks, characterized in that the receiving node has a virtual path switching device for synchronizing and switching the working and protection virtual paths using the control command code and the sequence number. . 9. A virtual path switching device used in the virtual path switching system according to claim 8, wherein a buffer memory that stores cells to be switched, which cells belong to the virtual path, and the buffer. A virtual path switching device, comprising: buffer control means for managing writing and reading of memory for each virtual path of a plurality of VPs. 10. The buffer memory is provided for each of two virtual paths, and the buffer control means reads out the same number of working and spare cells that have arrived after the same sequence number of the switching control OAM cell at the same time. 10. The selection means for selecting one of the read working and spare cells when the control command code of the switching control OAM cell is a switching request. Virtual path switching device. 11. A control means for controlling each virtual path temporarily stores a write address register provided for each virtual path to be managed for temporarily storing a write address given to the buffer memory, and a read address given to the buffer memory. A read address register for storing, a free address FIFO memory for storing a free address in the buffer memory, an address link memory having an address structure uniquely corresponding to a cell storage address in the buffer memory, and an arrived cell When storing in the buffer memory, the address is read from the write address register corresponding to the virtual path to which the arriving cell belongs, the arriving cell is stored in the address in the buffer memory, and the vacant address FIFO memory stores the vacant address. When reading, storing the empty address in the write address register, storing the empty address in the address in the address link memory, and reading the cell from the buffer memory, the address from the read address register corresponding to the virtual path to be read Is read out, the cell stored at the address in the buffer memory and the next address stored in the address link memory are read, and the next address is stored in the free address FIFO memory, and the next address is read out. 10. The virtual path switching device according to claim 9, further comprising an access control unit that stores it in a register.