JP3257858B2 - Network systems and switching nodes - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a plurality of switching nodes arranged in a mesh of N.times.M (N and M are integers of 2 or more) and grouped by each row and each column.
Communication is performed in one hop between any switching nodes in the same group, and communication is performed in two hops between any switching nodes belonging to different groups via switching nodes located at the intersection of the groups to which each switching node belongs. The invention relates to a network system configured to be performed and a switching node used for this network.

[0002]

2. Description of the Related Art In recent years, a network system capable of high-speed switching has been developed in a packet-switching type network in accordance with an increase in the speed of terminals and a connection request between local area networks.

[0003] In such a network system, it is required that even if the data amount increases, the exchange efficiency does not decrease due to data collision. Also,
It is required that the data arrival time does not vary depending on the positions of the source node and the destination node.

In order to meet this demand, the present applicant has proposed the following high-speed distributed network system in Japanese Patent Application No. 244626 filed on Sep. 14, 1992.

That is, in this network system, a plurality of high-speed packet data switching nodes are N × M
When communication is performed between arbitrary switching nodes, there is provided a group switch configuration of N columns and M rows in which the vertical direction (column) and the horizontal direction (row) constitute one group. Each switching node belongs to two groups, row and column.

When each communication node belongs to the same group as the group to which the own node belongs, each switching node is provided with a unit capable of communicating with this communication partner in one unit of communication time (hop).

If the communication partner belongs to a group different from the group to which the own node belongs, the communication partner belongs to one of the two groups to which the target switching node belongs, and one of the two groups to which the own node belongs. Means for relaying once at the switching nodes belonging to the group (there are also two) and communicating from the relay node to the target node at the next hop.

In other words, relaying is performed by one of two switching nodes located at two points forming another diagonal of a square having the communication node and the communication partner as diagonals, and the relay node switches from the relay node to the target node in the next time unit. Means are provided for communication.

According to such a configuration, when the communication partner belongs to the same group as the group to which the own node belongs,
Packet data can be transmitted with one hop, and when belonging to different groups, transmission can be performed with two hops, so that the variation in the arrival time of packet data can be reduced.

[0010] Even in the case of relaying a data packet, since only one relay node is required, even if the data amount increases, data collision hardly occurs. As a result, a decrease in transfer efficiency can be prevented.

[0011]

However, in the above-mentioned network system, since the grouping of the switching nodes is fixed, there is a problem that the transfer capability inherent in the system may not be fully utilized. Was.

That is, when the switching node A is added,
The additional node A does not always exchange data uniformly with all switching nodes on the network, and may exchange data only with a specific switching node B frequently.

For example, when a workstation is added in a LAN, this workstation is more effective than when exchanging data with another workstation.
More often, data is exchanged with a file server.

In the case where the extension node A has such a characteristic, if the extension node A belongs to a different group from the specific switching node B, the transfer form of the extension node A is two-hop transmission.

However, in the two-hop transfer, the transfer time is longer than in the one-hop transfer, so that the transfer efficiency deteriorates. In the two-hop transfer, the transmitting node waits for a signal indicating that the transfer has been normally completed from the relay node to return, and after confirming that this signal has been returned, transmits the next data. It must be output.

Therefore, when an exchange node A having the above-described characteristics is added, it is necessary to include this additional node A in the same group as a specific exchange node B.

However, in the conventional network system, the grouping of the switching nodes is fixed, so that it is difficult to realize the above-described connection form.

Thus, in the conventional network system, when the transfer mode is set to 2-hop transfer,
Although one-hop transfer is possible, there is a problem that this cannot be fully utilized.

Such a problem occurs not only when the switching node A is added, but also when, for example, the job processed by the terminal X under the switching node A changes.

That is, until now, a certain switching node A has frequently performed data transfer with the switching node B. However, another switching node A under the change of the job processed by the terminal X under the switching node A causes another switching. It is often necessary to perform data transfer with the node C.

In such a case, if the data transfer between the switching nodes A and B is one hop and the data transfer between the switching nodes A and C is two hops, the change of the processing of the terminal X causes the switching node A to receive two hops. Hop forwarding increases. As a result, also in this case, the transfer efficiency at the switching node A decreases.

As described above, in the conventional network system, since the grouping of switching nodes is fixed, the number of switching nodes operating on the network, the number of switching nodes at each time, and the connection to the switching nodes are changed. Depending on the usage status of the terminal, two-hop transfer is increased compared to one-hop transfer, and there is a problem that the transfer capability inherent in the network system cannot be fully utilized.

In other words, in the conventional network system, there is a problem that it is not possible to freely set an optimal and highly efficient node connection mode in consideration of traffic generated on the network.

SUMMARY OF THE INVENTION The present invention has been made to address the above circumstances, and provides a network system and an exchange node which can freely set an optimal and highly efficient node connection mode in consideration of traffic generated on a network. The purpose is to provide.

[0025]

In order to achieve the above object, the present invention has the following configuration in the above-described network system.

1. By using a frequency multiplexing technique as a signal transmission technique, all switching nodes are connected to a common transmission path.

2. The transmission means of each switching node is configured such that its transmission frequency is defined based on the group to which the own node belongs and can be changed according to the change of this group.

3. The receiving means of each switching node is configured to be able to receive signals from all switching nodes belonging to the group to which the own node belongs,
The receiving frequency can be changed according to the change of the group to which the own node belongs.

[0029]

In the above arrangement, the transmission means of each switching node outputs a signal in accordance with the transmission frequency specified by the vertical (column) and horizontal (row) groups to which the own node belongs. This transmission signal is supplied to all switching nodes via a transmission line shared by all switching nodes.

In the receiving means of each switching node, of the signals transmitted via the transmission line, the signals are transmitted from the switching nodes belonging to the vertical (column) and horizontal (row) groups to which the own node belongs. Only received signals are received. If the signal is a signal addressed to the own node, the signal is taken in. If the signal is a signal to be relayed, the signal is relayed.

In such a state, when the group to which the own node belongs is changed, the transmission frequency of the transmission means and the reception frequency of the reception means are changed based on the change destination group.

As a result, the group to which each switching node belongs can be changed, so that an optimal and highly efficient node connection mode in consideration of traffic generated on the network can be set freely.

[0033]

Embodiments of the present invention will be described below in detail with reference to the drawings.

In the following description, a case where the present invention is applied to a network system using an optical frequency multiplexing technology as a frequency multiplexing technology will be described as a representative.

1. First Embodiment First, a first embodiment of the present invention will be described.

Logical Arrangement of Switching Nodes FIG. 1 is a diagram showing a logical arrangement of switching nodes in a network system to which the present invention is applied.

As shown, the switching node is N × M (N,
M is a positive integer of 2 or more). In addition, each switching node has a horizontal direction (row) and a vertical direction (co
(lumn). Here, the group of the n-th (n = 1, 2,..., N) row is row-n, and the group of the m-th (m = 1, 2,..., M) column is
column-m. Each node has a node number, which represents the coordinate position of the switching node in the network. For example, the node number (n, m) indicates that the switching node is located at n rows and m columns in the network.

Each switching node receives only transmission data of switching nodes belonging to the same group. When performing packet transfer, 1 is set between switching nodes belonging to the same group.
Data is transferred at the hop. On the other hand, data is transferred in two hops between switching nodes belonging to different groups via nodes that perform a relay operation.

The physical connection diagram 2 of the switching node is a diagram showing the physical topology of the transmission path of each switching node.

As shown in the figure, the physical connection form of each switching node can be any of a bus type, a ring type, and a star type. In this embodiment, separate lines are used for the optical transmission line in the horizontal direction (row) and the vertical direction (column). That is, two optical transmission lines are used.

FIGS. 3A and 3B show a physical connection form of switching nodes in a bus type. In the figure, 10
0 indicates an optical transmission line in the horizontal direction (row) constituted by optical fibers and optical couplers. Hereinafter, this optical transmission line 100
It is called a row bus. Reference numeral 101 denotes a vertical (column) optical transmission line composed of an optical fiber and an optical coupler. The following optical transmission path 101 is called a column bus.

As shown in the figure, each of the buses 100 and 101 is commonly connected to all the switching nodes because the optical frequency multiplexing technique is used as the signal transmission technique.

Optical Frequency Assignment Method The transmission frequency of each switching node has a unique value defined by the coordinate position of its own node. In other words, the node has a unique value defined by the group of the horizontal direction (row) and the vertical direction (column) to which the own node belongs. Here, the transmission frequency of the switching node having the node number (n, m) is represented by f (n, m).
Is defined. Here, for example, it is assumed that frequency allocation is performed using the periodic characteristics of the Fabry-Perot filter.

The Fabry-Perot type filter is shown in FIG.
As shown in (1), the transmission band (BW) has a characteristic that appears periodically (period: FSR) on the frequency (freq) axis.

The center frequency of one of these transmission bands (BW) is f (1,1), the center frequency one cycle higher than f (1,2) is f (1,2), and the center frequency is one cycle higher than that. The frequencies are assigned by the number of nodes M included in row-1, such that the center frequency is f (1,3),. This is referred to as a row-1 frequency group (row (1)).

Next, as shown in FIG. 4 (b), using the characteristics obtained by shifting the characteristics of this filter to the high frequency side, f (2,1) to f (2, M) are used in the same manner as above. Assign to At this time, the assignment is made such that f (2,1) comes immediately above the high frequency of f (1,1). This is called row-2 frequency group (ro
w (2)). By repeating this, frequencies are allocated by the number N of row groups. Generally, a frequency group assigned to row-n is referred to as a row-n frequency group (row (n)).

FIG. 5 shows a diagram in which frequency allocation has been performed. In the figure, the frequency of the solid line is assigned to row-1, the dashed-dotted line is assigned to row-2, and so on. In this figure, the frequencies from f (1,1) to f (N, 1) belong to column-1, so the column-1 frequency group (colu
mn (1)). Similarly, the frequencies from f (1,2) to f (N, 2) are defined as column-2 frequency groups (column (2)),.
The frequencies from (1, M) to f (N, M) are defined as a column-M frequency group (column (M)). In general, f (1, m) to f
Frequency up to (N, m) is column-m frequency group (column (m)
).

By making such an assignment, the horizontal direction
An arbitrary frequency can be selected by adopting a two-stage configuration of a filter for selecting (row) and a filter for selecting vertical direction (column). The allocated frequency is required for the number of nodes (N × M).

The diagram 6 of switching node shows a structure of a transceiver of the switching node (n, m).

As shown in the figure, the transmitting / receiving section is composed of an optical signal transmitter 200 and an optical signal receiver 300.

The transmission frequency of the optical signal transmitter 200 is set to a unique value defined by the position of the own node as described above. That is, the own node is the switching node (n, m)
, F (n, m) is assigned as the transmission frequency. The light emitting element of the optical signal transmitter 200 has, for example, a variable wavelength capable of appropriately setting emission wavelengths (f (1,1) to f (N, M)) for N × M channels by changing a bias voltage. It is composed of a laser diode.

The optical signal output from the optical signal transmitter 200 is applied to a row bus 100 and a column bus 101, respectively.
Transmitted to all switching nodes.

The optical signal receiver 300 includes a horizontal (row) receiving section and a vertical (column) receiving section.
The receiver in the horizontal direction (row) includes a tunable wavelength filter 301, an M-branch coupler 302, and M fixed filters 303 (1).
To 303 (M). The receiving section in the vertical direction (column) also includes a tunable wavelength filter 311, an N
It is composed of N fixed filters 313 (1) to 313 (N).

Here, the variable wavelength filter 301 in the horizontal direction (row) extracts an optical signal of row (n) from the input light of the row bus 100. That is, the optical signal from the switching node belonging to the row-n to which the own node belongs is extracted. The variable wavelength filter 301 can appropriately set the passing wavelengths (row (1) to row (N)) for N channels by changing the bias voltage.

The M-branch coupler 302 divides the output light of the tunable wavelength filter 301 into M light beams and supplies them to the fixed filters 303 (1) to 303 (M).

The fixed filters 303 (1) to 303 (M) are
From the branch output of the M branch section 302, column (1) to co
Extract the light signal of lumn (M). As a result, optical signals having transmission frequencies f (n, 1) to f (n, M) appear at the output terminals of the fixed filters 303 (1) to 303 (M). That is, the optical signal from the switching node belonging to the row-n to which the own node belongs appears in a separated state.

The fixed filters 303 (1) to 303 (1) to 303
The output of (M) is supplied to a reception data processing unit (not shown) and undergoes a predetermined reception process. That is, if the received data is addressed to its own node, it undergoes decryption processing and the like. On the other hand, if the received data is data to be relayed, it undergoes relay processing for sending to the destination.

The variable wavelength filter 31 in the vertical direction (column)
1 extracts an optical signal of column (n) from the input light of the column bus 101. That is, column-n to which the own node belongs
Extract the optical signal from the switching node belonging to. By changing the bias voltage, the variable wavelength filter 311 transmits the wavelengths for M channels (column (1) to column (M)).
Can be appropriately set.

The N-branch coupler 312 branches the output light of the variable wavelength filter 311 into N lights, and supplies them to the fixed filters 313 (1) to 313 (N), respectively.

The fixed filters 313 (1) to 313 (N) are
From the branch output of the N-branch coupler 312, row (1) to ro
Extract the optical signal of w (M). As a result, optical signals having transmission frequencies f (n, 1) to f (n, M) appear at the output terminals of the fixed filters 303 (1) to 303 (M). That is, the optical signal from the switching node belonging to column-m to which the own node belongs appears in a separated state.

The fixed filters 313 (1) to 313 (313)
The output of N) is also supplied to a reception data processing unit (not shown) and undergoes predetermined reception processing.

As described above, in this embodiment, N + M + 2 filters are used in the receiving section of each switching node. In this case, the fixed filter that passes the transmission frequency of the own node is omitted.

Transmission Operation The optical signal of frequency f (n, m) output from the optical signal transmitter 200 is supplied to a row bus 100 and a column bus 101, respectively. Thereby, this optical signal is supplied to all switching nodes.

Reception Operation The input light from the row bus 100 is first supplied to the first-stage variable wavelength filter 301. As a result, an optical signal of the row-n frequency group (row (n)) is extracted. The extracted optical signal is branched into M pieces by an M-branch coupler 302, and then supplied to second-stage fixed filters 303 (1) to 303 (M). As a result, in the optical signal (multiplexed signal) of row (n), optical signals of frequencies f (n, 1) to f (n, m) are separated.

Similarly, the input light from the column bus 101
First, an optical signal of column (m) is extracted by the first-stage variable wavelength filter 311. Next, this optical signal is N-branched by an N-branch coupler 312 and then the second-stage fixed filter 31.
At 3 (1) to 313 (N), optical signals of frequencies f (1, m) to f (N, m) are separated.

Rearrangement of Switching Nodes For example, consider a case where a switching node having a node number (1,1) is moved to a switching node having a node number (2,3).

In this case, the transmission frequency of the optical signal transmitter 200 is changed from f (1,1) to f (2,3). In the optical signal receiver 300, the pass band of the tunable wavelength filter 301 is ro.
w (1) is changed to row (2), and tunable wavelength filter 3
Eleven pass bands are changed from columns (1) to columns (3). These changes are achieved by slightly changing the bias voltage of the filter or the like and shifting the frequency.

Effects According to this embodiment described in detail above, the following effects can be obtained.

(A) First, the transmission frequency of the optical signal transmitter 200 and the reception frequency of the optical signal receiver 300 can be changed, so that the physical connection form of the switching node is not changed and the logical connection form is free. Can be changed to

As a result, it is possible to freely set an optimal and efficient node connection mode in consideration of traffic generated on the network. Hereinafter, a specific example thereof will be described.

(A) For example, when the transfer between the switching nodes is two hops, the two-hop transfer can be changed to the one-hop transfer by changing the logical position of the transmission side or the reception side. Thereby, the transfer efficiency between nodes can be improved.

(B) If the switching node fails for some reason and 2-hop transfer via this node becomes impossible, the logical positions of the normal node and the failed node are exchanged to obtain 2 hops. The transfer can be continued.

(C) Further, it is now assumed that one computer is connected to the switching node and the switching node and the computer form an integrated device. When adding or removing such communication devices,
Depending on the number of communication devices connected on the network,
The connection form between the communication devices can be optimized.

(B) Further, since the optical frequency is allocated using a wavelength filter having a periodic characteristic, such as a Fabry-Perot type filter, a wavelength filter having no such characteristic is used. The number of wavelength filters in the switching node is reduced as compared with the case where allocation is performed.

For example, although the row and column groups each have N × M different frequencies, each switching node has a two-stage configuration of a filter for selecting a row and a filter for selecting a column. So N + M +
Only two filters are required. Further, since the amount of frequency shift of the reception filter unit is not known, it is easy to control this.

2. Second Embodiment Next, a second embodiment of the present invention will be described.

Configuration This embodiment is characterized by the configuration of the optical signal receiver 300 of the switching node.

That is, in the first embodiment, the case where the variable wavelength filter is used as the first-stage filter and the fixed filter is used as the second-stage filter has been described. On the other hand, in this embodiment, a variable wavelength filter is used for both the first and second stages.

FIG. 7 shows the configuration of the optical signal receiver 300 in this case. In the figure, 304 is an M-1 branch coupler,
314 is an N-1 branch coupler, 305 (1) to 305 (M-1)
, 315 (1) to 315 (N-1) are variable wavelength filters in the second stage.

The logical arrangement of the switching nodes, the physical connection form, the allocation of optical frequencies, and the configuration of the optical signal transmitter are as follows.
This is the same as that of the first embodiment.

Rearrangement of switching node For example, consider the case where switching node (1,1) is moved to the position of switching node (2,3). Also in this case, the optical signal transmitter 20
The transmission frequency of 0 is changed from f (1,1) to f (2,3).

On the other hand, in the optical signal receiver 300, the pass band of the first-stage variable wavelength filter 301 on the horizontal (row) side is changed from row (1) to row (2), and Tunable filter 315 (3) has a pass band from column (3)
Changed to column (1). Also, the pass band of the first-stage variable wavelength filter 311 in the vertical direction (column) is set to columns (1).
Is changed to column (3), and the pass band of the second-stage variable wavelength filter 315 (2) is changed from row (2) to row (1).

Effects In this embodiment, the same effects as those of the first embodiment can be obtained, and the filters 303 (m) and 313 (n) that pass the transmission frequency of the own node can be omitted. Therefore, an effect is obtained that the number of filters can be reduced to N + M.

3. Third Embodiment Next, a third embodiment of the present invention will be described.

Configuration This embodiment is characterized in how to provide an optical transmission line.

That is, in the first and second embodiments, the case where the optical transmission lines are separately provided in the vertical direction (row) and the horizontal direction (column) has been described. On the other hand, in this embodiment, the vertical direction (ro
The optical transmission line is shared by w) and the horizontal direction (column). In this case, the allocated frequency needs to be twice the number of nodes (N × M × 2).

FIGS. 8A and 8B show a physical connection mode of switching nodes in a bus type. In the figure, 10
Reference numeral 3 denotes an optical transmission line composed of an optical fiber and an optical coupler, which is shared in a horizontal direction (row) and a vertical direction (column). Hereinafter, the optical transmission path 103 is referred to as a bus.

Optical frequency allocation The optical frequency allocation is shown in FIGS. 9 (a) and 9 (b).

The difference from the allocation in the first and second embodiments is that the bus 103 in the horizontal direction (row) and the vertical direction (column) is used.
The transmission frequency in the horizontal direction (row) and the vertical direction (column) is divided into different frequency band bundles (row-band, respectively).
column-band).

The transmission frequency of the switching node (n, m) in the row direction is defined as f (n, m, r), and the frequency of f (n, 1, r) to f (n, M, r) The group is row (n, r) and the frequency group from f (1, m, r) to f (N, m, r) is colu
As mn (m, r), row-band frequency allocation is performed as shown in FIG.

On the other hand, the vertical direction (column) of the switching node (n, m)
F (n, m, c) and f (1, m, c) to f (N, m, c)
Column (m, c) and f (n, M, c) to f (n, M, c) as row (n, c), as shown in FIG.
Assigns -band frequency.

Configuration and operation of switching node FIG. 10 shows the configuration of the transmitting / receiving section of the switching node.

The optical signal transmitting section 200 frequency multiplexes the transmission signal in the horizontal direction (row) and the transmission signal in the vertical direction (column) and sends out the signal to the bus 103. That is, the switching node (n, m
) Optical signal transmitter 200 transmits the transmission signal of f (n, m, r) and f (n, m, r)
The transmission signal of (n, m, c) is multiplexed and transmitted. In this case, a variable wavelength laser diode capable of changing N × M × 2 channels is used as the light emitting element.

The optical signal receiver 300 has a three-stage filter configuration. The first-stage filters 331 and 341 are fixed filters that pass row-band and column-band optical signals, respectively. The filters at the second and third stages have the same configuration as the filters at the first and second stages in FIG. Therefore, these are denoted by the same reference numerals as in FIG.

In the above configuration, the input light from the bus 102 is branched into two by the two-branch coupler 321 and then supplied to the row-band pass filter 331 and the column-band pass filter 341 respectively.

As a result, the row-band pass filter 331
, A row-band optical signal is extracted. This optical signal is supplied to the variable wavelength filter 301 at the next stage. Thereby, the optical signal of row (n, r) is extracted. This optical signal is M-branched by the M-branch coupler 302,
Fixed filter 30 that passes optical signals of (1, r) to column (M, r)
3 (1) to 303 (M). This allows row-
The optical signals of the band are separated into optical signals of f (n, 1) to f (n, M).

Similarly, the column-band pass filter 341
Extracts the column-band optical signal. This optical signal is supplied to the variable wavelength filter 311 at the next stage. Thereby, the optical signal of column (m, c) is extracted. After this optical signal is M-branched by the M-branch coupler 312,
fixed filter 313 that passes optical signals of row (1, r) to row (M, r)
(1) is supplied to 313 (M). This gives the column
The -band optical signal is separated into optical signals of f (1, m) to f (N, m).

Rearrangement of Exchange Node For example, consider a case where an exchange node (1,1) is moved to an exchange node (2,3). In this case, the transmission frequency of the optical signal transmitter 200 is f (1,1, r) + f (1,1, c) to f (2,3, r) + f (2,3, c).
Is changed to In the optical signal receiver 300, the pass band of the variable wavelength filter 301 is changed from row (1, r) to row (2, r).
And the pass band of the filter 303 of the variable wavelength filter 311 is changed from column (1, c) to column (3, c).

[0099] It is possible to obtain the same effect as the first embodiment, of course, further, the effect of the optical transmission line requires only one is obtained.

4. Fourth Embodiment Next, a fourth embodiment of the present invention will be described.

Switching Node Configuration This embodiment is different from the third embodiment in that the optical signal receiver 3
As in the second embodiment, a variable wavelength filter is used as the second and third filters of 00. FIG. 11 shows the configuration. 11, the same parts as those in FIGS. 7 and 10 are denoted by the same reference numerals.

The input light from the receiving operation bus 103 is split into two by a two-branch coupler 321 and then split into a row-band pass filter 331 and a
It is supplied to a lumn-band pass filter 341.

Thus, the row-band pass filter 331
, A row-band optical signal is extracted. This optical signal is supplied to the variable wavelength filter 301 at the next stage. Thereby, the optical signal of row (n, r) is extracted. This optical signal is M-1 branched by an M-1 branch coupler 302, and then column (1, r),..., Column (m-1, r), column (m + 1,
r),... fixed filter 305 that passes optical signals of column (M, r)
(1) to (305) (M-1). This allows the row-ba
The optical signals of nd are f (n, 1), ..., f (n, m-1), f (n, m + 1), ... (n,
M).

Similarly, the column-band pass filter 341
Extracts the column-band optical signal. This optical signal is supplied to the variable wavelength filter 311 at the next stage. Thereby, the optical signal of column (m, c) is extracted. This optical signal is N-1 branched by an N-1 branch coupler 314,
Row (1, c), ... row (n-1, c) row (n + 1, c), ...,
fixed filter 315 (1) to 31 for passing the optical signal of row (N, c)
5 (N-1). Thus, the column-band optical signals are f (1, m), ..., f (n-1, m), f (n + 1, m), ..., f (N,
m).

Rearrangement of Exchange Node For example, consider a case where an exchange node (1,1) is moved to an exchange node (2,3). In this case, the transmission frequency of the optical signal transmitter 200 is f (1,1, r) + f (1,1, c) to f (2,3, r) + f (2,3, c).
Is changed to

In the optical signal receiver 300, the pass band of the filter 304 in the second stage in the horizontal direction (row) is changed from row (1, r) to row
(2, r), the third-stage variable wavelength filter 305 (3)
Is changed from column (3, r) to column (1, r).

The fixed filter 314 in the second stage in the vertical direction (column) is changed from column (1, c) to column (3, c), and the pass band of the fixed filter 315 (2) in the third stage is changed. row
(2, c) is changed to row (1, c).

Effects According to such a configuration, the same effects as in the second embodiment can be obtained in the configuration as in the third embodiment.

In the above description, the case where the present invention is applied to an optical communication network system has been described. However, the present invention can also be applied to a telecommunication network system.

In addition, it goes without saying that the present invention can be variously modified and implemented without departing from the scope of the invention.

[0111]

As described in detail above, according to the present invention,
It is possible to provide a network system and a switching node that can freely set an optimal and highly efficient node connection mode in consideration of traffic generated on a network.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a logical arrangement of switching nodes according to a first embodiment;

FIG. 2 is a diagram illustrating a physical connection mode of switching nodes according to the first embodiment;

FIG. 3 is a diagram illustrating a physical connection mode of switching nodes in the bus-type connection according to the first embodiment;

FIG. 4 is a diagram illustrating frequency characteristics of a Fabry-Perot filter;

FIG. 5 is a diagram illustrating a frequency allocation state according to the first embodiment.

FIG. 6 is a block diagram illustrating a configuration of a switching node according to the first embodiment;

FIG. 7 is a block diagram illustrating a configuration of a switching node according to a second embodiment;

FIG. 8 is a diagram illustrating a physical connection mode of switching nodes in a bus-type connection according to a third embodiment;

FIG. 9 is a diagram illustrating a frequency allocation state according to the third embodiment.

FIG. 10 is a block diagram illustrating a configuration of a switching node according to a third embodiment;

FIG. 11 is a block diagram illustrating a configuration of a switching node according to a fourth embodiment.

[Explanation of symbols]

100 row bus, 101 column bus, 102 bus, 200 optical signal transmitter, 300 optical signal receiver, 3
01, 311, 305 (1)-305 (M-1), 315 (1)
To 315 (N-1) ... variable wavelength filter, 302 ... M branch coupler, 303 (1) to 303 (M), 313 (1) to 313 (N)
... fixed filter, 304 ... M-1 branch coupler, 314
... N-1 branch coupler, 312 ... N branch coupler, 321 ...
2-branch coupler.

──────────────────────────────────────────────────続 き Continuing on the front page (72) Ryuichi Watanabe 1-7-12 Toranomon, Minato-ku, Tokyo Oki Electric Industry Co., Ltd. (56) Reference JP-A-6-97941 (JP, A) , Et al., Study on application of optical technology to grid switching network, IEICE Technical Report, December 17, 1992, SSE92-110 (58) Fields investigated (Int. Cl. ⁷ , DB name) H04L 12 / 56 H04B 10/20 H04L 12/28

Claims (2)

(57) [Claims] A plurality of switching nodes are N × M (N and M are 2
Are arranged in a mesh shape of the above integers) and are grouped for each row and each column, communication is performed at one hop between any switching nodes in the same group, and between any switching nodes belonging to different groups. In a network system configured so that communication is performed in two hops via a switching node located at the intersection of a group to which each switching node belongs, a transmission path shared by all switching nodes, The transmission frequency is defined based on the group, the transmission means is capable of changing the transmission frequency according to the change of the group to which the own node belongs, and receives the signal sent from the switching node belonging to the group to which the own node belongs. Receiving means capable of changing the receiving frequency according to the change of the group to which the own node belongs. Network system, characterized in that. 2. A method according to claim 1, wherein the plurality of switching nodes are N × M (N and M are 2
Are arranged in a mesh shape of the above integers) and are grouped for each row and each column, communication is performed at one hop between any switching nodes in the same group, and between any switching nodes belonging to different groups. In the switching node of the network system configured to perform two-hop communication via the switching node located at the intersection of the group to which each switching node belongs, the switching node connected to the transmission line shared by all the switching nodes. A transmission means capable of changing the transmission frequency defined based on the group to which the own node belongs according to the change of the group to which the own node belongs, and a transmission line shared by all the switching nodes, and It receives signals sent from switching nodes belonging to the group to which it belongs, and sets the reception frequency to Switching node, wherein the node is equipped with a receiving means capable of changing in accordance with changes of the group to which it belongs.