JPH0738641B2 - Communication network control method and apparatus - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a communication network control method and apparatus, and more specifically, a local area network (hereinafter referred to as LAN).
Control method and device.

2. Description of the Related Art Conventionally, communication network control methods and devices include, for example, (a) CSMA baseband LAN and (b) broadband LA.
N, (c) TDMA baseband LAN and digital PBX,
(D) "Optical communication network" described in Japanese Patent Application Laid-Open No. 57-104339, which is the prior application of the present invention, and (e) "Communication described in Japanese Patent Application Laid-Open No. 58-139543, which is also the prior application of the present invention. Circuit ",
(F) Similarly, there is a “communication network control system” described in Japanese Patent Application No. 60-170428, which is a prior application of the present invention. However, the CSMA baseband LAN of (a) above is suitable when the packet is short like data information or text information and suddenly occurs, but as in multimedia communication, the packet is infinite. When the length is long, continuous occurrence occurs, or real-time performance is required, there is a problem that communication collisions occur frequently and a high throughput (communication capacity) cannot be obtained. Multimedia communication in LAN includes image information, audio information, and video information in addition to conventional text communication of data, and is a combination thereof. In addition, (b) Broadband LAN has a shortage of capacity for multimedia communication.
There are problems with cost and scalability. Further, (c) the TDMA (time division) baseband LAN and the digital PBX are most suitable for multimedia communication among the conventional methods, but there are problems in cost and expandability. Especially, when it is used as multimedia communication, the cost becomes very high. further,
(D) JP-A-57-104339 (hereinafter referred to as the first invention), JP-A-58-139543 (hereinafter referred to as the second invention)
Each of the methods described in the publication is the basis of the present invention and is most suitable for multimedia communication. To briefly explain these prior inventions, the first invention is characterized by the principle of first-come-first-served logic, multiple-coupling structure, and path fixing by multiple-input one-output elements. That is, as shown in FIG. 18-1, a large number of terminals 21 connected via the communication path 52 are provided.
The communication network is composed of a to d, and each terminal that communicates is the 14th
By sending out four basic packets shown by 60 to 63 in the figure, each node is controlled and communication is performed. Each node 50
a to d have a multi-bond structure, and the order in which any of the packets 60 to 63 in FIG. 14 arrives, that is, by passing only one communication by the first-come-first-served logic and waiting the other , Multi-input / single-output operation is realized. Then, each node fixes one communication path by changing the sending direction so as to sequentially reverse the transfer direction of the series of packets shown in 60 to 63 of FIG. According to the first aspect of the invention, each node is provided with a function of detecting a destination address, and each node detects the destination address added to the packet, so that if it is addressed to the terminal connected to its own node, , Output it to the terminal. If it is not addressed to the terminal of its own node, the destination address is not used for any control. Reference numeral 60 in FIG. 14 is a call packet, which has a destination preamble part, a destination address part, and a message part. However, since the call packet is a packet that only calls the other party, there is usually no content in the message section. Reference numeral 61 denotes a call bucket packet, which only needs to have a preamble added, and the called terminal returns this call bucket packet when the destination address matches its own address. Each node through which the call packet has passed reverses the transfer direction after a lapse of a predetermined time, so that the call back packet can reach the caller terminal in the reverse direction along the route through which the call packet has passed. Reference numeral 62 denotes a message packet, and the packet to which the preamble and the message have been added is transferred from the calling party terminal to the called party terminal. That is, the transfer direction of each node through which the call back packet has passed reverses after a lapse of a predetermined time, so that the message packet can reach the called party terminal accurately even if no address is added.
Reference numeral 62 is an ACK packet, and the called terminal that has correctly received the message sends this packet to the calling terminal. Each node through which the message packet has passed reverses the transfer direction after a lapse of a predetermined time, so that the node can accurately reach the caller terminal even if no address is added. In this way, the path is fixed by simply adding the address to the first packet, and each packet can make two round trips.

Next, the second invention is basically the same as the first invention, but by degrading the function of each node, the node cannot detect the address, and all the addresses are detected by the terminal. Is.

As described above, in each of the inventions of the prior applications, when there is a multi-input / single-output, that is, when there is a transmission request from one terminal to one node, the node permits only the first request by the first-come first-served logic. Only one terminal makes a call transfer. Therefore, the connection control means of the node 50 cannot simultaneously perform a plurality of communications in parallel. However, this is extremely inconvenient. For example, in FIG. 18-1, if a path having a specific pattern as shown by a black line is currently fixed between the terminal 51a and the terminal 51b, the path is fixed. Available links to the three nodes through which
Even if 52 is left, if there is a first-arrival input signal for one link, the remaining links will be disconnected and the network will be divided into left and right. For example, from terminal 51c to terminal 51d. Even when you want to communicate with
There is an inconvenience that the connection cannot be made until the communication between 51a and 51b is completed. The solution to this problem is
(F) The method described in Japanese Patent Application No. 60-170428 (hereinafter referred to as the third invention). The present invention enables a single node to handle a plurality of communications at the same time, fixing a specific pattern path so as not to hinder other communications, and enabling large capacity communications by effectively utilizing links. It is something to do. That is, as shown in FIG. 18-2,
Since each node 50 can handle multiple communications, for example, a terminal can be used without being affected by the pattern of fixed paths.
Communication between 51g and terminal 51h, communication between terminal 51i and terminal 51j, and communication between terminal 51f and terminal 51e can be performed in parallel. In the third invention, in any node, the first
When the transmission of the forward information and the first return information is completed, all the other channels are put in the connected state and only the channel having the earliest input signal is output. However, in the third invention, since a plurality of nodes are provided, a large scale
It is effective as a LAN, but it is a small L with multiple terminals.
There is a problem that it cannot be applied as AN.

OBJECT The object of the present invention is to improve such a problem, to be able to handle a plurality of communications at the same time, and to be effective as a small-scale communication network whose network is composed of only one node, and to have a large network with the same protocol. An object of the present invention is to provide a communication network control method and device which can be expanded to a large scale LAN.

Configuration To achieve the above object, a communication network control method of the present invention has a single node and a plurality of transmission / reception terminals connected to the node, and the node has a plurality of input channels and the input. A connection control means for controlling the connection state with the output channel corresponding to the channel is provided, and the connection means selects only the input channel in which the first outgoing information arrives earliest as the entire output channel or the output channel corresponding to the input channel. When all the except output channels are connected and the input channels other than the input channels are disconnected from the output channels, and the transmission of the first outgoing information of the first incoming input channel is completed, the input channel is supported. The input channel corresponding to the output channel that has output the first outgoing information to the output channel is connected to pass the first outgoing information. In the communication network control method, the input channel corresponding to the output channel of the first recovery information is connected to the output channel corresponding to the input channel of the first recovery information repeatedly. The control means detects the input channel having the first input signal earliest, stores the input channel in the first, and outputs the first input signal to the input channel corresponding to the output channel which outputs the first input signal. When there are two input signals, the second input signal is detected and the second storage is performed, and the second input signal is output from the output channel corresponding to the input channel stored first. , Above first
And if there is a third input signal in a channel other than the second stored input channel, the input channel that received the third input signal earliest is the second input signal that is the first input signal and its recovered signal. The input signal was connected to all output channels except each channel, the input channel having the earliest third input signal of the input signals was stored in the third, and the third input signal was output. When the input channel corresponding to the output channel has the fourth input signal, the input channel is detected and stored in the fourth, and the output channel corresponding to the input channel in which the fourth input signal is stored in the third It is characterized by repeating the operation of outputting from. Further, the connection control means, all output channels that are not already used for other communication, that is, not in the path fixed state,
Alternatively, if the input signals are connected to all the output channels that are not in the path fixed state except the output channel corresponding to the input channel and there are input signals in the plurality of input channels, the input signal is once output from the output channel, and then the The input channel of the first outgoing information of the first communication of the first arrival is detected and stored first, and the connection to the output of the input channel other than the stored input channel is disconnected, Upon completion, after connecting the input channel corresponding to the output channel that outputs the forward information to the output channel corresponding to the first stored input channel, the input for which the forward information of the first communication is input It is also characterized in that the channel is stored second and a fixed path state is formed.

Further, in the communication network control device of the present invention, the connection control means is a switching matrix device including a switching element array for simultaneously connecting an arbitrary input channel to an arbitrary output channel in a plurality of combinations, and an input signal is first input. A first-come-first-served input signal detection device that detects existing input channels, an input signal monitoring device that outputs to the outside when the information regarding the presence or absence of input signals for all input channels changes, and A control gate device consisting of a switching element for connecting only the input channel to the first-arrival input signal detecting device, and reading information about the input signal from the first-arrival input signal detecting device and the input signal monitoring device, Switches for switching element arrays and control gate devices And a switching control device for controlling the switching element array.

Hereinafter, the constitution of the present invention will be described with reference to examples.

Prior to the embodiments, the principle of the present invention will be described in comparison with the conventional one.

FIG. 3 is a connection configuration diagram of a conventional optical star type network. In FIG. 3, 101a to 101d are terminals, and 102 is a star coupler. Since the physical topology of the network of this embodiment is a star type, it will be explained by comparison with a conventional optical star type network. Although the physical form of the optical star type network is a star type, it is easier to understand the connection form by expressing the star type from the side as shown in FIG. Each transmitting / receiving terminal 101a-d
The input port IN and the output port OUT of are connected to the output port OUT and the input port IN of the star coupler 102, respectively. FIG. 4 is a diagram showing an access method of the optical star type network of FIG. The packet sent from the transmitting terminal 101a passes through the route indicated by the thick line to the star coupler 102.
Input from input port IN and output port OUT from each terminal 1
It is output to the line connected to 01b-d. in this way,
The state that the packet from the transmitting terminal spreads during the network is shown, and it can be seen that the logical topology is the bus type. Usually, because of competition for this one bus, the access method is set to CSMA / CD (Carier sence mult).
iple access wuth collision detect) is used.

5 and 6 are block diagrams of the star coupler in FIG. The node shown in FIG. 5 has a number of O / Es (Optical to Electronic conv) equal to the number of input / output channels.
ert) and E / O (Electronic to Optical convert). The node shown in FIG. 6 has a number of O / Es equal to the input / output channels and one E with a large output energy. It is a configuration with / O. In addition to these, there is also a configuration in which the number of O / Es is set to one by using high-sensitivity O / E or by combining the optical fibers of the input channel into one by a directional coupler. In FIGS. 5 and 6, 110 is an optical fiber, 111 is a photodiode / transistor, 112 is an inverter, 113 is a MOS transistor, and 114 is a LE.
D, 115 light fiber.

In the star coupler 102, for example, packets from four terminals are transmitted through the four optical fibers 110, converted into electric signals by the corresponding four photodiodes 111 at the input port, and then combined into one line. . And
In the output port of FIG. 5, the inverter 112 operates the LED driving transistor 113 to send the light emitted from the four LEDs 114 to all the terminals 101a to 101d via the corresponding four optical fibers 115. . Further, in the output port of FIG. 6, one transistor 113 is operated by the inverter 112 and the light emitted from one LED 114 is received by the four optical fibers 115 and sent to each of the terminals 101a to 101d. .

FIG. 1 is a functional explanatory diagram of a communication network control method showing an embodiment of the present invention. In this embodiment, only one node 120 is provided in the network, and this node 120 controls communication from a plurality of terminals 101, 104 to 106. In FIG. 1, 101a to 101e are first to fifth terminal devices,
For example, it is a word processor having a keyboard display or a communication function, or a personal computer. Further, 104 is a disk server, 105 is an optical disk server, and 106 is a print server. Here, the first terminal device 101a is outputting to the print server 106, and the second terminal device 101b is reading the file from the optical disk server. In this embodiment, as described above, in one node 120, a plurality of communications can be controlled simultaneously in parallel.

FIG. 2 is an explanatory diagram of a connection control method for the nodes in FIG. 2, the same symbols as in FIG. 1 indicate the same devices. The node 120 does not distinguish between node-to-node connection and node-to-terminal connection, and has no control over increasing the number of nodes. Further, even if the number of nodes is increased and the system is expanded, it is not necessary to change the protocol.

In FIG. 2A, the node 120 establishes a fixed link between the first terminal 101a and the print server 106 by passing the forward information and the backward information. Then the second
In the figure (b), after the first fixed link setting, the first outgoing information (call packet) from the second terminal 101b that received the call first is sent to the remaining terminals 101c-e, 104-105. . In FIG. 2C, the communication direction is reversed at the node 120 after a lapse of a fixed time after the first outgoing information (call packet) has passed. In FIG. 2D, only one piece of the first recovery information (call back packet from the destination terminal 104) is passed. This recovery information reaches the transmitting terminal 101b even if the destination address is not added. In FIG. 2E, fixed links for the first information communication (communication between the terminal 101a and the print server 106) and the second information communication (communication between the terminal 101b and the disk server 104) are set. That is, two separate communications are possible at one node 120 at the same time.

Next, the detailed configuration of the node 120 in FIG. 1 will be described. FIG. 7 is an internal block diagram of the node in FIG. As shown in FIG. 7, the node 120 includes input ports i0 to i7, output ports o0 to o7, a switching matrix device 1 for interconnecting the corresponding input channels and output channels, and a first-come-first-served logic. To the first-arrival input signal detecting device 2 for detecting an input channel for which the first-arriving input signal is received, an input signal monitoring device 3 for constantly monitoring the presence or absence of an input signal, and an arbitrary input channel for the first-arrival input signal detecting device 2. Control gate device 4 to be connected and switching control device 5 for controlling the entire node including them
It has and. Assuming that the input channels are 8 channels, the switching matrix device 1 is composed of 8 modules divided for each output channel.
The first-arrival input signal detection device 2, the input signal monitoring device 3, and the control gate device 4 each comprise one module. The switching control device 5 is connected to these modules by a module select bus 6, a gate set bus 7, and a data bus (module output signal line) 8. That is, the switching control device 5
Receives information from the first-arrival input signal detecting device 2 and the input signal monitoring device 3 through the data bus 8, and selects each module of the switching matrix device 1 through the module select bus 6 and inputs the first-arriving input signal detecting device 2 into the module. Controls the signal monitoring device 3 and gate / set bus 7
The switching matrix device 1 and the control gate 4 are controlled by.

8-1 to 8-8 are diagrams showing an operation sequence of each device of FIG. 7. FIG. 8-1 is a diagram showing the initial state of the node 120. In the initial state, the switching control device 5 controls all the input channels of the control gate 4 by the control signal to detect the first-arrival input signal. The input channel i0 of the switching matrix device 1 is connected to the device 2 and controlled by a control signal.
~ I7 is interconnected to output channels o0-o7.

FIG. 8-2 is a diagram showing a state in which an input signal is received, in which one of the input channels (here, the input channel).
If there is an input signal at i2), that input signal is output from all output channels o0 to o7. On the other hand, the input signal of the input channel i2 input to the control gate 4 is also transferred to the first-arrival input signal detection device 2. Further, the input signal is also input to the input signal monitoring device 3.

FIG. 8-3 is a diagram showing a case where an input signal is received also in another input channel, and the first-arrival input signal detection device 2 detects that the input signal is received earliest in the input channel i2,
The switching control device 5 reads out that effect by the control signal, and performs the first storage. Then, if there is an input signal in the other input channels as well, they are output from the output channels o0 to o7 in the same manner. At this time, interference occurs in the output signal.

FIG. 8-4 is a diagram in the case of disconnecting the connection to the output channels other than the earliest input channel, and the switching control device 5 uses the control signal to input signals other than the input channel i2 of the switching matrix device 1. Input channel i.e. input channel i0 ~ i0, i3 ~ i7 output channel o0 ~ o7
Disconnect to.

FIG. 8-5 is a diagram showing a modified operation of FIG. 8-4, in which the earliest input signal is prevented from being connected to the output channel corresponding to the input channel, or a link having a fault is provided. It is also possible not to connect to the output channel of. To this end, the terminal device has a specific configuration such that when the input channel has the first outgoing information, the terminal device outputs the first outgoing information from the output channel.
That is, the switching control device 5 connects the input channel i2 of the switching matrix device 1 to all output channels to transmit a signal.
Even if the output to the output channel corresponding to i2 is returned to the original terminal, it is useless except for confirmation, so the connection to the output channel corresponding to the input channel is cut off. Further, even when the signal is sent from this node to the adjacent node, or when the first outgoing information is sent to the terminal device having the specific configuration, the same signal should be returned immediately. , If there is an incoming channel (for example, an input channel i6) that has not been returned, consider that the link has failed or the adjacent node or terminal is not powered and the input channel i6 The corresponding output channel o6 can be disconnected.
This disconnects the connection to the specific output channel by controlling with the control signal from the switching controller 5.

FIG. 8-6 is a diagram in the case where the transfer direction is switched in preparation for the forward information after the forward information is finished. The input signal monitoring device 3 detects that the input signal is finished, and the switching control device is shown. 5 detects this by the control signal and controls the switching matrix device 1 by the control signal, whereby the input channels i0 to i7 or the input channels i0 to i1, i3 to i7 or the input channels i0 to i1, ( i2), i
Connect 3 to i5, i7 to the output channel o2. By this, even if the return information from any input channel is input,
It can be returned to the sender of the outgoing information.

FIG. 8-7 is a diagram showing the input state of the return information, and when the return information for the forward information transmitted previously is input to any of the input channels (input channel i0 in the figure) within a predetermined time. , This information is output from the already connected output channel o2.

When the first communication is formed by the forward information sent last time and the returning information sent this time, the input signal monitoring device 3 inputs the input signal of the first returning information of the first communication to the input channel i0. Is detected, the switching control device 5 reads it out by the control signal, performs the second storage, and receives the control signal by the input channel i1, i of the control gate 4.
3 to i7 are connected to the first-arrival input signal detecting device 2 to reset the first-arrival input signal detecting device 2 by a control signal, and the input channels i1, i3 to i7 of the switching matrix device 1 are output by the control signal. Interconnect to ~ o7.
This waits for input signals from all the remaining input channels except the input channels i0 and i2 and the output channels o0 and o2 that are currently communicating.

FIG. 8-8 is a diagram showing a case where a plurality of communications are simultaneously handled, in which an input signal from the input channel i0 is output to the output channel.
While connected to o2, the other input channel (in the figure,
When there is an input signal from the input channel i4), it is output from all output channels except o0 and o2 during communication as before. Of course, the output channels during communication and the output channels corresponding to the input channels corresponding to the input signals may be connected to all the output channels. In the subsequent processing, first, when the first forward information of the second communication is completed, the input channel corresponding to the output channel is processed to the input channel (in the figure, the input channel i4) which receives the input signal. Connect to the output channel (o4) and wait for the input signal. Further, if the signal of the first communication is completed first, the transfer direction of the input / output channel is switched.

FIG. 9 is a block diagram of the switching matrix device 1 in FIG. The case where each of the input channels and the output channels is 8 channels will be described. Each module is composed of 8 modules, and each module has 8 switching gates 10 and 8 latches connected to them. And the output of switching gate 10
When 10c is input, it is composed of one 8-input OR element 12. Each module shares the gate set bus 7 with the eight input signal lines from the input ports. Latch 11
The D terminal 11a is connected to the gate set bus 7, the Q terminal 11c is connected to the input terminal 10b of the switching gate 10, and the G terminal string 11b is connected to the module select bus 6 by the common enable line 11b '. The input terminal 10a of the switching gate 10 is connected to the input signal line, and the output terminal 10c is 8
It is connected to the input of the input OR element 12.

FIG. 10 is a block diagram of the control gate device in FIG. The case where the input channel is an 8-channel will be described. The gate is composed of 8 gates 40 and 8 latches 41 connected thereto. Eight input signal lines from the input port and the corresponding first-arrival input signal detection device 2
Control the connection of 8 output signal lines to.

The D terminal 41a of the latch 41 is connected to the gate / set bus 7 by Q
The terminal 41c is connected to the input terminal 40b of the gate 40, and the G terminal row
41b is connected to the module select bus 6 by a common enable line 41b '. The input terminal 40a of the gate 40 is connected to the input signal line, and the output terminal 40c is connected to the input of the first-arrival input signal detection device 2.

FIG. 11 is a logical structure diagram of a latch used in the switching matrix device 1 of FIG. 9 and the control gate device 4 of FIG.

These are composed of two NAND gates having a switching function in the front stage and two NAND gates having a storage function in the rear stage.

The configurations of FIGS. 9 to 11 are not limited to this, and there are many design variations. Deformation occurs primarily due to the ratio of software to hard wafer used to perform the aforementioned controls.

In this embodiment, since the switching control device 5 is composed of a microprocessor, the control of this part uses software, and the switching matrix device 1 and the control gate device 4 also function correspondingly. Independent. If the above control is performed only by hardware, these devices are functionally inseparable and integrally configured.

FIG. 13 is a configuration diagram of a network when the present invention is expanded.

In principle, the network of the present invention is a small-scale LAN that includes only one node 120, but the number of nodes can be increased in sequence to expand to a large-scale network. The network to be expanded may be of an indeterminate shape, and may be linear, loop, secondary lattice type as shown in FIG. 13, three-dimensional lattice type, or a combination thereof. And between any nodes 50, nodes
A link 52 including a plurality of channels can be connected between the terminal 50 and the terminal 51. In the above-described embodiments, one link 52 is described as including at least one input channel and one output channel, but it is of course possible to use a plurality of channels or one input / output channel. Is.

FIG. 14 is a block diagram of a packet used according to the present invention. In the packet structure according to the present invention, restrictions on the communication procedure required for the transmitting / receiving station such as a terminal are defined as follows.

(A) The first outgoing information (call packet) 60 is provided with a preamble 60a having a predetermined length (time) or more and an intended receiving terminal address (test address) 60b.

(B) The transmitting / receiving terminal receives the first outgoing information (call packet) 60 addressed to itself, and when it ends, immediately after the elapse of the first voting time (T1) 64, the first returning information. (Call back packet) 61 is transmitted. First predetermined time (T1)
64 is a time required for the connection control means of the node to perform control for inputting the first recovery information 61, and is called a node time constant (node constant).

(C) When the transmitting / receiving terminal receives information that is not addressed to itself (only the first outgoing information is received), the transmitting / receiving terminal performs transmission within the second predetermined time (T2) 65 after the end of the information. Don't get it. The second predetermined time (T2) 65 is the time required for the packet to propagate through the network, and is called the network time constant (network constant). This allows
Even if there are a plurality of nodes, the node closest to the transmitting terminal has the first recovery information (call back packet) 61 within the second predetermined time (T2) 65 after the end of the first forward information (call back packet) 60. Guaranteed to have input.

As long as the restrictions on the communication procedure described above are complied with, other degrees of freedom are wide, and the following is possible.

(A) There is no limit on the maximum packet length. (B) There is no limit to the number of times the forward information and the backward information can be continuously repeated, and the channel can be monopolized. (C) The data rate can be freely determined between the transmitting and receiving terminals as long as it is less than or equal to the maximum data rate determined by the hardware that constitutes the network. The packet shown in FIG. 14 is the most general packet structure, and consists of two pieces of forward information and two pieces of backward information.
The first outgoing information 60 and the first returning information 61 do not include a message because they are used to secure a communication path on the network and release an unnecessary portion to another.

FIG. 15 is a block diagram of the first-arrival input signal detection device in FIG. The case where the input channel is 8 channels will be described. The first-arrival input signal detection device 2 is composed of eight latches 20 and eight rear gates 21. The latch 20 turns off all the gates 20b when there is an input signal 20a in any of them,
It is output from the output 20d, which is restored by the clear signal 20c. The D terminal row 20a is connected to the output of the control gate device 4, and the CLR terminal row 20c is connected to the module select bus 6 by one common clear line 20c '. Q terminal row
20d is connected to one of the two input terminals 21a of the rear gate 21, respectively. The other input terminal 21b of the rear gate 21 is connected to the module select bus 5 by one common enable line 21b ', the output terminal 21c is connected to the data bus 5, and the output terminal 21c is connected to the data bus (module Output signal line) 8.

FIG. 12 is a logical structure diagram of the latch of FIG. The two front stage NAND gates in FIG. 12 have a switching function, and the two rear stage NAND gates have a storage function.

FIG. 16 is a block diagram of the input signal monitoring apparatus in FIG. The case where the input channel is 8 channels will be described.
It consists of eight latches 30 and eight post-stage gates 31.
The D terminal 30a of the latch 30 is connected to the input port and the Q terminal 3
0d is connected via an inverter 32 to the G terminal 30b of the latch and one input terminal 31a of the rear gate 31. The output terminal 31c of the rear gate 31 is connected to the data bus 8. The CLR terminal 30c of the latch 30 is connected to the module select bus 6 by a common clear line 30c ', and the other input terminal 31b of the latter stage gate is connected to the module select bus 6 by a common enable line 31b'.

The switching controller 5 selects one module of the switching matrix device 1 by the module select bus 6 and sets the gate connection state by the gate set bus 7. Further, the module select bus 6 reads information from the first-arrival input signal detection device 2 or the input signal monitoring device 3 and clears the latch.
The module select bus 6 and the gate set bus 7 set the connection state of the gate of the control gate device 4.

Next, the operation of the node showing another embodiment of the present invention will be described with reference to FIGS. 17-1 to 17-11.

In FIG. 17, the structure of the matrix of the switching matrix device 1 is partially modified so that the intersections on the diagonal lines are not connected and the input channels are not connected to the corresponding output channels.

By the way, when the switching matrix device 1 is composed of N × N switching elements, it is possible to connect the input channel to the output channels other than the output channel corresponding to the input channel before the input signal is input. is there. In addition, the switching matrix device 1 is
When configured with a switching element of × (N-1), the input channel is inherently not connected to the output channel corresponding to the input channel. Therefore, when the switching matrix device 1 having such a structure is used,
The operation sequence shown in FIG. 8 is the sequence shown in FIG. In addition, in FIG. 17, the matrix of the switching matrix device 1 is 8 × 8 and the diagonal lines are not connected, but an 8 × 7 matrix may be used.

FIG. 17-1 shows a state corresponding to FIG. 8-1, that is, an initial state. The switching control device 5 controls the switching matrix device 1 and the control gate device 4 by a control signal to connect all the input channels to the first-arrival input signal detection device 2, and the input channel of the switching matrix device 1 is controlled. i0 to i7 are all output channels
O0 to o7 are interconnected except on the diagonal.

FIG. 17-2 shows a state corresponding to FIG. 8-2, that is, a state where an input signal is received. Now, input channel i2
Input signal to the output channel o0, o
Output from 1, o3 to o7. Signals are also input to the first-arrival input signal detection device 2 and the input signal monitoring device 3.

Fig. 17-3 shows a state corresponding to Fig. 8-3, that is, a state in which another input signal is input. If an input channel i2 is subsequently input to input channels i1 and i3, they are also input. All the output channels except the output channel corresponding to the input channel are output. At this time, interference occurs in the output signal and interference also in the input signal.

FIG. 17-4 shows the state corresponding to FIG. 8-3, that is, the detection operation of the earliest input channel when an input signal is input to another input channel. Following the input channel i2, there are input signals at i1 and i3, and further i0, i4 to i5, i7
If there is an input signal, the switching matrix device 1 once connects those input channels to all output channels except the output channels corresponding to the input channels, and outputs them. In this case, it is not output because it is not connected to the output channel corresponding to the input channel. The first-arrival input signal detection device 2 detects the earliest-arrival input channel i2, and the switching control device 5 reads it out to make the first
Is remembering.

FIGS. 17-5 and 17-6 each show a state corresponding to FIG. 8-4, that is, an operation state in which the input channel other than the earliest input channel is disconnected. The switching control device 5 controls the switching matrix device 1 by reading the first-arrival input channel i2 from the first-arrival input signal detection device 2 and storing the input channel i2, that is, the input channels other than the input channel i2, that is, the input channels i0 to i. Disconnect the output channels o0 to o7 of i1, i3 to i5, i7.

17-7 and 17-8 each show the state of FIG. 8-6, that is, the direction switching operation after the earliest-arrival input signal is completed. When the input signal monitoring device 3 detects that the signal of the input channel i2 has ended, the switching control device 5 reads this and controls the switching matrix device 1 by the control signal.
Connect the input channels i0, i1, i3 to i7 to the output channel o2. That is, by switching the input channel that was previously communicated to the output channel, no matter which input channel the back information is input to, the back information is returned to the transmitting terminal.

FIGS. 17-9 and 17-10 each show a state corresponding to FIG. 8-7, that is, a state of inputting backward information. If the previous input signal is the forward information forming the first communication,
The restoration information of the first communication is input to one of the input channels within a predetermined time (T1). Here is the input channel
Since the return information has been input to i0, this is output to the output channel o2. The input signal monitoring device 3 detects that the input signal of the first recovery information of the first communication is input to the input channel i0, and the switching control device 5 respectively reads and performs the second storage (control signal ). Next, the control gate device 4 is controlled to connect the input channels i1, i3 to i7 of the gate device 4 to the first-arrival input signal detecting device 2, and the first-arrival input signal detecting device 2 is reset to obtain the last-first-arrival input signal. Erase the channel. As a result, the information of the second communication and the information of the third communication can be input, and a plurality of communication can be handled at the same time.

17-11 shows the state corresponding to FIG. 8-8, that is, the simultaneous handling state of plural communications.

As the second communication, the output channel corresponding to the input channel of the output channels o0 and o2 during a call is output.

FIG. 19 is a block diagram of a node showing another embodiment of the present invention. In the embodiment of FIG. 7, it is assumed that each link 52 includes a plurality of channels, and it has been described as including at least one input channel and one output channel. However, when the link 52 includes only one input / output channel, as shown in FIG. 19, a bidirectional driver 72 is connected to one input / output channel 71 to connect with the input port of the switching control device 5 of the node. Connect to output port. When eight links are connected to one node, eight driver 72 are also connected.

FIG. 20 is a block diagram of the bidirectional driver (RS422) in FIG. A and B terminals are connected to the link side, and D (driver), R (receiver), and DE (driver enable) and RE (receiver enable) are connected to the node side, respectively. The positive polarity signal sent from the driver D is connected to the terminal A and the reverse polarity signal is connected to the terminal B, and the terminal extracts the signal by the voltage difference between the two terminals A and B. On the other hand, the terminal D is connected to the output port and the terminal R is connected to the input port, so that both transmission and reception are not enabled at the same time.
DE and RE are controlled by the switching controller 5.

Advantageous Effects As described above, according to the present invention, one node can handle a plurality of communications, and the cost is hardly increased. It can handle even large-scale ones.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a system showing an embodiment of the present invention, FIG. 2 is an explanatory diagram of a connection operation in FIG. 1, FIGS.
The figure shows the structure and operation of an optical star type network for comparison with the present invention, FIGS. 5 and 6 are the block diagrams of the star coupler in FIG. 3, and FIG. 7 is the node of FIG. FIG. 8 is an internal configuration diagram, FIG. 8 is a diagram showing the operation sequence of FIG. 7, FIG. 9 is a configuration diagram of the switching matrix device in FIG. 7, and FIG. 10 is a configuration diagram of the control gate device in FIG. FIG. 11 is a block diagram of the latch in FIGS. 9 and 10, FIG. 12 is a block diagram of the latch in FIG. 15, and FIG. 13 is a block diagram of the network at the time of expansion of the present invention.
FIG. 14 is a block diagram of a packet used in the present invention, FIG. 15 is a block diagram of the first-arrival input signal detecting device in FIG. 7, FIG. 16 is a block diagram of the input signal monitoring device in FIG. 7, and FIG. Is a motion sequence chart showing another embodiment of the present invention,
FIG. 18 is a block diagram of a conventional network, FIG. 19 is a block diagram of a node showing another embodiment of the present invention, and FIG. 20 is a block diagram of the bidirectional driver in FIG. 101,104,105,10
6: Terminal, 120: Node, 102: Star coupler, 1: Switching matrix device, 2: First-arrival input signal detection device, 3: Input signal monitoring device, 4: Control gate device, 5: Switching control device, 6: Module Select Bus, 7: Gate Set Bus, 8: Data Bus, 50: Node, 51: Terminal, 52:
Link.

Claims (6)

[Claims] 1. A single node and a plurality of transmission / reception terminals connected to the node, wherein the node controls connection states of a plurality of input channels and output channels corresponding to the input channels. Connection control means is provided for connecting only the input channel receiving the earliest first forward information to all output channels or all output channels except the output channel corresponding to the input channel, and When the connection of the input channel other than the input channel to the output channel is disconnected and the transmission of the first incoming information of the first incoming input channel is completed, the first outgoing information is output to the output channel corresponding to the input channel. The input channel corresponding to the output channel is set to the connection state, the first outgoing information is passed, and the output channel of the first backward information is output by the end of the first outgoing information. In the communication network control method which repeatedly operates so as to connect the corresponding input channel to the output channel corresponding to the input channel of the first decoded information, the connection control means is the earliest to receive the first input signal. The first input signal is stored in the first input channel, and the second input signal is stored in the input channel corresponding to the output channel that outputs the first input signal. The signal is detected, the second storage is performed, the second input signal is output from the output channel corresponding to the first stored input channel, and the first storage is performed.
And if there is a third input signal in a channel other than the second stored input channel, the input channel having the third input signal earliest is used as the first input signal and the second input which is the recovery signal. An output that is connected to all output channels except each channel of the signal, stores the input channel having the earliest third input signal of the input signals third, and outputs the third input signal. When the input channel corresponding to the channel has a fourth input signal, the input channel is detected and stored in the fourth, and the fourth input signal is output from the output channel corresponding to the third stored input channel. A communication network control method characterized by repeating the operation of outputting. 2. The first input signal is the first of the first communication.
Forward information of the first communication is the first return information of the first communication, and the second forward information of the first communication subsequent thereto is the output corresponding to the input channel of the second storage. The second communication information of the first communication is output from the output channel corresponding to the input channel of the first storage, while the third input signal is the first input of the second communication. The forward information, the fourth input signal is the first return information of the second communication, and the second forward information of the subsequent second communication is the output channel corresponding to the input channel of the fourth memory. 2. The communication network control method according to claim 1, wherein the second recovery information of the second communication is output from an output channel corresponding to the input channel of the third memory. . 3. A single node and a plurality of transmission / reception terminals connected to the node, wherein the node controls connection states of a plurality of input channels and output channels corresponding to the input channels. Connection control means is provided for connecting only the input channel for which the first outgoing information arrives earliest to all output channels or all output channels except the output channel corresponding to the input channel, When the connection of the input channels other than the above to the output channels is cut off and the transmission of the first incoming information of the first incoming input channel is completed, the first outgoing information is output to the output channel corresponding to the input channel. The input channel corresponding to the output channel is set to the connected state, and the first return information as a response to the first forward information is output from the input channel in the connected state. In a communication network control method for repeatedly operating so as to connect to an output channel corresponding to an input channel of first outgoing information, the connection means is not already used for other communication, that is, is not in a path fixed state. If there is an input signal in multiple input channels by connecting to all output channels that are not in the path fixed state except the output channel or the output channel corresponding to the input channel, the input signal is once output from the output channel. Then, the input channel of the first outgoing information of the earliest first communication is detected and stored first, and the connection to the output of the input channel other than the stored input channel is disconnected, When the forward information is completed, the input channel corresponding to the output channel that outputs the forward information is set to the output channel corresponding to the first stored input channel. A communication network control method, characterized in that, after connecting to, the input channel for which the first communication forward information was input is stored in the second, and the path fixed state is formed. 4. A single node and a plurality of transmission / reception terminals connected to the node, wherein each node controls a connection state of a plurality of input channels and an output channel corresponding to the input channel. Connection control means for controlling the connection control means to connect only the input channel receiving the earliest first outgoing information to all output channels or all output channels except output channels corresponding to the input channel, and When the connection of the input channels other than the input channels to the output channels is disconnected and the transmission of the first incoming information of the first incoming input channel is completed, the first outgoing information is sent to the output channel corresponding to the input channel. The input channel corresponding to the output channel that has been output is placed in the connected state, the first forward information is passed, and the output channel of the first backward information is output when the first forward information is completed. In the communication network control device which repeatedly operates so as to connect the input channel corresponding to the input channel to the output channel corresponding to the input channel of the first decoded information, the connection control means sets the arbitrary input channel to the arbitrary input channel. A switching matrix device consisting of a switching element array for connecting to output channels in multiple combinations at the same time, a first-come-first-served input signal detection device that detects the input channel that first received an input signal, and the presence or absence of input signals for all input channels. Control signal consisting of an input signal monitoring device that outputs to the outside when there is a change in information, and a switching element that connects only an arbitrary input channel of a plurality of input channels to the first-arrival input signal detection device. And the input signal from the first-arrival input signal detection device and input signal monitoring device. Reads information, communication network control apparatus, characterized in that it comprises a switching control unit for controlling the switching element array of the switching element array and the control gate device of the switching matrix apparatus. 5. The switching matrix device according to claim 1, wherein when the number of input channels and the number of output channels of the switching element array are the same, N, N.times.N elements are provided at all the combination intersections of the input channels and the output channels. The communication network control device according to claim 4, wherein the communication network control device comprises: 6. The switching matrix device according to claim 1, wherein when the number of input channels and the number of output channels of the switching element array are N, the corresponding input channel and output channel among all combinations of input channels and output channels. 5. The communication network control device according to claim 4, wherein the communication network control device is composed of N × (N−1) elements provided at the combination intersections except for.