JP3395703B2 - Point-to-point communication network system and communication control method therefor - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a point-to-point communication network system and its control method,
In particular, the present invention relates to a learning table control method in a communication network system configured by a plurality of communication stations that perform point-to-point communication with both adjacent stations and refers to a learning table to make a communication data routing decision.

[0002]

2. Description of the Related Art The spread of LANs (Local Area Networks) has become more widespread and the purpose of their use has become more sophisticated, and the number of communication systems for connecting a plurality of LANs to each other is increasing. In such a communication system, a bridge device is required to connect a plurality of LANs, and these bridge devices require a learning table (learning function table) to manage the location of the terminal device on the network of the communication system. become. The details of such a learning table are disclosed in, for example, Japanese Patent No. 2746246 and are well known.

[0003]

Conventionally, in a network configured by a learning bridge having such a learning table, a function of filtering communication data by the learning table is required, but the maintenance of the learning table is communicated. Since the data itself is used, there is a drawback that when the route configuration dynamically changes, an error occurs in the routing or the information in the learning table cannot be updated quickly. Further, when the communication path includes a closed loop configuration, a solution such as a spanning tree algorithm is necessary to avoid a data storm of broadcast data, and complicated bridge information exchange is required between all bridges.

An object of the present invention is to provide a point-to-point communication network system and a communication control method for the same so that the learning table can be automatically and quickly updated even if the communication path configuration dynamically changes. That is.

Another object of the present invention is to provide a point-to-point communication network system capable of preventing a data storm of broadcast data with a simple structure and a communication control method thereof even when the communication path includes a closed loop structure. That is.

[0006]

According to the present invention, there is provided a communication network system comprising a plurality of communication stations which perform point-to-point communication with both adjacent stations and make a routing decision of communication data by referring to a learning table. Each of the communication stations is a means for exchanging a request message and a response message for confirming the connection between the adjacent stations, and whether each of the communication stations is an endpoint station according to whether the response message is received or not. Means for determining whether or not, and when it is determined to be the end point station, issue a route information notification for indicating a communication route with the end point station as its own address, and broadcast it,
When it is determined to be a station other than that, means for inserting the own station address in the received route information notification and carrying out broadcast propagation to the next station, and updating the contents of the learning table according to the received route information notification. And a communication network system.

The communication station determined to be the terminal station is characterized in that it controls the end of the route information notification and the end of all broadcast data in addition to the generation of the route information notification. The sending and receiving of the response message and the issuance of the route information notification are both performed at a constant cycle.

Furthermore, the communication station refers to the route information notification to determine whether or not the own station exists in the closed loop, and if it is determined that the own station exists in the closed loop, the communication station Means for determining whether or not it should be a closed-loop end point station, and when it is determined to be the closed-loop end point station, in addition to the generation of the route information notification, to control its end and the end of all broadcast data. It is characterized by having done.

According to the present invention, there is provided a communication control method in a communication network system configured by a plurality of communication stations which perform point-to-point communication with both adjacent stations and make a routing decision of communication data by referring to a learning table. Then, in each of the communication stations, the step of exchanging a request message for confirming the connection and its response message between adjacent stations, and whether or not the own station is an end point station according to the presence or absence of the response message of And if it is determined to be the end point station, it issues a route information notification indicating the communication route with the end point station as its own address and broadcasts it, and it is determined to be another station. In this case, the step of inserting the own station address into the received route information notification and carrying out the broadcast propagation to the next station, and performing the broadcast according to the received route information notification. A communication control method characterized by including the step of updating the contents of the packaging table can be obtained.

The communication station determined to be the terminal station is characterized in that, in addition to the generation of the route information notification, it controls the end of the route information notification and the end of all broadcast data. The transmission and reception of the response message and the issuance of the route information notification are both performed in a fixed cycle.

Further, in each of the communication stations, a step of judging whether or not the own station exists in the closed loop by referring to the route information notification, and a step of judging if the own station exists in the closed loop, Determining whether or not the station should be a closed-loop end point station, and in the case of determining that the station is the closed-loop end point station, in addition to the generation of the route information notification, managing its end and the end of all broadcast data. It is characterized by doing so.

The operation of the present invention will be described. In a point-to-point communication network consisting of digital communication stations that perform point-to-point communication with neighboring left and right communication stations, each communication station has a learning bridge function and the location of all communication stations connected to the network. The following functions are provided in each communication station when the learning table stores which side is left or right with respect to its own communication station.

That is, a health check message (a request message for confirming the connection between the adjacent stations and a response message thereof) is transmitted and received to and from the adjacent communication station, and it is judged from this result whether or not the terminal station is the terminal station. Function and, if it is judged to be an end point station, issue a route information notification to show the communication path with the end point station as its own address and broadcast it, and receive it if it is judged to be another station A function to insert the own station address in the route information notification that has been made and broadcast to the next station,
A function of updating the contents of the learning table according to the received route information notification is provided. Then, the communication station determined to be the terminal station is provided with a function of controlling the end of the route information notification and the end of all broadcast data.

Further, by referring to the route information notification, it is judged whether or not the own station exists in the closed loop. If it is judged that the own station exists in the closed loop, it is judged whether or not the own station should be the closed loop end point station. If a closed-loop terminal station is determined, it is possible to control the end of the route information notification and the end of all broadcast data. With such a configuration, even if the network includes a branch or a closed loop, and its form dynamically changes due to a network failure,
The routing route can be corrected automatically and quickly.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a schematic network configuration of a point-to-point digital communication network to which the present invention is applied. In FIG. 1, each communication station on the network has communication ports in two directions, a wireless system and a terrestrial LAN system, and performs point-to-point communication with an adjacent communication station. Further, since each communication station also performs port-to-port data transfer, it is possible to communicate with all communication stations connected to the network via adjacent communication stations. Furthermore, the wireless system connection in this network is only one-to-one communication, but in the terrestrial LAN system connection, it is possible to connect between multiple communication stations, so the network branches via the LAN or loops. It is also assumed that they are formed in.

In this configuration, each communication station has an address (MAC (Media Access Control)) as station identification information.
Addresses) are respectively assigned, and each of these communication stations operates as a learning bridge. Each communication station has a learning table 1 as shown in FIG.
1 is provided, and when the data is transferred, the learning table is referred to for routing and filtering the transfer data to prevent the transfer data from being unnecessarily transferred to the network. The learning table 11 needs to be automatically and promptly updated so that an error does not occur in routing when the route configuration dynamically changes due to a failure of the communication network.

Therefore, each communication station 10 is configured as shown in FIG. In FIG. 2, each communication station 10 has two ports 1 and 2, and the data transmission / reception unit 15 controls the data transmission / reception between the ports 1 and 2 while referring to the learning table 11. This learning table 11
A control unit (CPU) 14 is provided for the management control of the device, an adjacent station information table 12 and a closed loop information table 13 are provided, and the CPU 14 is exchanged via ports 1 and 2 (described later). The adjacent station information table 12 and the closed loop information table 13 are generated according to the contents of the health check message and the route information notification message (shown in FIGS. 7 and 12, respectively). When the own station exists in the closed loop, the control unit 14 refers to the tables 12 and 13 to determine whether the own station is the closed loop end point station. Further, the control unit 14 controls updating of the learning table 15 according to the content of the route information notification message.

The learning table 11 has the structure shown in FIG. 3. Basically, all the information registered in the learning table 11 is updated only by the route information notification shown in FIG. However, normal data will not be updated. Furthermore, even if there is a closed loop configuration on the network and there are multiple routes that reach any one communication station, only one address information regarding that communication station exists in the learning table and double registration is not performed. .

The route information regarding the communication station is the wireless port /
Only which side of the terrestrial port exists is an important factor, and it is sufficient if only the MAC address of each communication station is written in the learning table information, but there are multiple routes for any communication station. In this case, the number of hops to each communication station is also written in the learning table for the purpose of selecting and registering a port that becomes a short route (in a certain communication station, the route information for an arbitrary communication station may arrive from both ports. , It is only immediately after the network configuration has changed, and normally information of any communication station is received only from the same port direction. Route information notification may arrive). Registration time information is also written in the learning table for the purpose of deleting old route information.

FIG. 4 shows the contents of the adjacent station information table 12, showing which communication stations are connected to the ports 1 and 2, for example, in the case of the network shown in FIG. In the existing branch station, since a plurality of communication stations (two in FIG. 9) are connected to the branch port,
The MAC addresses of these communication stations are stored. The storage update of the adjacent station table 12 is performed by each communication station periodically transmitting and receiving the health check message shown in FIG. 7 (details will be described later).

FIG. 5 is a diagram showing the contents of the closed loop information table 13. The closed loop information table 1 is received according to the contents of the route information notification shown in FIG. 12A, which is periodically issued by the communication station that has become the terminal station.
3 is generated, and each MAC address of a communication station that forms a closed loop including its own station is stored.

Next, the operation of the embodiment of the present invention will be described. FIG. 6 is an operation flow of each communication station, and is a processing flow for determining whether or not the own station is an end point station. In order to monitor the configuration of the network directly connected to each communication station and the failure status, each communication station periodically communicates with the adjacent communication stations.
The health check messages shown in (A) and (B) are exchanged (step S1), and the MAC address of the adjacent communication station that has made a response is stored in the adjacent station information table 12 shown in FIG. 4 of itself (step S2). ).

Regarding this health check, considering the change of the branch route and the adjacent communication station, the health check message is notified by broadcast notification only for one hop. In consideration of the case where the neighboring communication station dynamically changes, the destination address of the health check is set to the broadcast address without fixing the partner address, but the receiving station of the health check transfers the relevant health check message to another port. I don't.

The response to the above health check message is returned for the number of communication stations adjacent to the communication station, but the health check message is sent from any one of the two ports of the wireless system / terrestrial LAN system at any communication station. If there is no response or request (step S3), this communication station is defined as the end point of the network (step S4). The definition of this terminal station is not fixed but dynamic. When a route failure occurs at a certain point, the communication stations on both sides of it become terminal stations. On the contrary, when the route failure is recovered or the route is extended, the terminal station disappears and operates as a normal communication station (step S6). The above processing is periodically performed at regular intervals (step S5).

FIG. 8A shows an end point station in a simple linear network topology. In FIG. 8, the communication stations (addresses A and E) at both ends of the linear network are defined as terminal stations. The numbers 1 and 2 in the figure indicate port numbers (the same applies hereinafter). The adjacent station information table 12 obtained in each communication station in this case is as shown in FIG.

When the network topology includes the branch as shown in FIG. 9A, the plurality of health check response messages are returned from one port. In this case, the communication station directly connected to the branch route is defined as a branch station, and the branch stations (each of the addresses C, D, and G) store all the addresses having the health check response in the adjacent station information table 12. . The state of the adjacent station information table 12 of each communication station at this time is shown in FIG. 9 (B).

The communication station which has become the terminal station executes the processing flow shown in FIG. That is, the communication station which has become the terminal station periodically transmits the route information notification shown in FIG. 12A to the adjacent communication station by broadcasting (steps S11 and S12). This route information notification is an infinite hop notification by broadcast and the propagation is stopped by the end point station. The end point station type and the end point station MAC address are written in the route information notification. The terminal station type information is used to identify whether the terminal station is a normal terminal station or a closed loop terminal station on a closed loop (described later).

The communication station which has received the route information notification is shown in FIG.
It operates according to the processing flow shown in. That is, each communication station that has received the route information notification (step S22) updates its own learning table 11 (step S22).
23), the MAC address of its own station is added to the route information notification, and the same is transmitted to the port on the opposite side of the port that received the route information notification (step S24). The broadcast of the route information notification is stopped by the communication station that has become the terminal station (steps S23 and S25). Note that FIG. 12B shows an example of the contents of the route information notification in the linear network topology of FIG. 8A. The end point type indicates the type of end point station, and there are a perfect end point and a closed loop end point. In the example of this linear network topology, both end point stations A and E are perfect end points. Incidentally, FIG.
2 (B) shows the contents of the route information notification transmitted from the perfect endpoint station A.

When the network route has a closed loop configuration as shown in FIG. 13A, there is no communication station which is an end point station. Further, in the closed loop configuration as shown in FIG. 13B, there is a risk that the route information notification or general broadcast notification issued by the end point station circulates in the closed loop to generate a broadcast storm. Therefore, a closed-loop terminal station is defined as a special terminal station by the following procedure.

Here, the closed loop as shown in FIGS. 14A and 14B will be described as an example. The contents of the adjacent information table in each of these cases are as shown in FIGS. 15A and 15B, respectively.
The procedure for determining a closed-loop end point station as a special end point station in these cases is shown as a flow in FIG. In the cases of FIGS. 14A and 14B described above, since the perfect end point station A is present in both cases, it is assumed that there is no normal change in the path configuration and there is no dynamic change in the route configuration. , FIG. 14
In (A) and (B), the perfect endpoint station A starts issuing the route information notification shown in FIG. 12 (A) and transmits it by broadcast (step S31). In the case of the simple closed loop as shown in FIG. 13, among the four communication stations that form the closed loop, the communication station that detects the earliest that there is no path information notification starts issuing the path information notification. To do.

Next, each station receives this route information notification and sequentially adds its own station address and broadcasts it to the next station.
Station B, station C, and station F in the case of FIG.
The route information notification as shown in (A) is received. As a result, the closed loop is detected (step S32). That is, the route information notification originating from the station A enters the loop from two directions at the branch of the station C and the station F, goes around the loop and returns to the branch point. Therefore, detecting a closed loop means
In the route information notification that was re-received after the route information notification circulated,
It means to detect that the address of the own station is already registered. Similarly, in the case of FIG. 14 (B), the stations that may detect the closed loop are A, B, and F, respectively, when the route information notifications as shown in FIG. 17 (B) are received.

Next, B, C, which detect these closed loops,
Station F (in the case of FIG. 14A), A, B, F (FIG. 14)
In the case of (B)), it is necessary for each one station to move to the closed loop end point station (step S33). That is, it is necessary to judge whether or not the next station exists in the closed loop, and if it is outside the closed loop, it is not possible to shift to the closed loop end point station.

First, the communication station which has detected the closed loop discards the received route information notification and determines whether it is inside or outside the closed loop. Taking FIG. 14A as an example, of the B, C, and F stations that have detected the closed loop, the B station receives the route information B>C>D>E>F> (B) and B>F> E. > D
>C> (B), a closed loop is detected. However, station B is a branch station in which a plurality of adjacent communication stations exist in port 2 of the adjacent station information table. Need to do.

This determination is made by "whether or not any two of the communication stations adjacent to the same port are included in the route information notification". In the case of station B, two stations C and F adjacent to port 2 are included in the route information. Therefore, station B itself represents either of the above (represents the same closed loop, but as route information It is also a station outside the closed loop with respect to the (two received) path information, and cannot be a closed loop terminal station.

On the other hand, in the case of station C, since it is a branch station having a plurality of adjacent stations (B, F) at port 1, the same judgment is made, but route notification C>D>E>F> (C ),
Of the (B, F) stations, since only one station, F station, is included, station C is determined to be a station in the closed loop and transitions to the closed loop end point station. Station C, which has become a closed-loop terminal station, will register closed-loop information in the closed-loop information table and will continue to issue route information notifications to both ports on a regular basis. Similarly, in the F station, the F station holds the following closed loop information table as a closed loop end point according to the determination as to whether it is inside or outside the closed loop, and periodically issues the route information notification.

The addresses registered in the closed loop information table are, as shown in FIGS. 18A and 18B, the looped portion of the addresses on the route information notification (from the address after itself). Address until the end).
Further, in the closed loop information table, the order of registered addresses is not important as long as the address group forming each closed loop can be recognized. For example, it can be seen that the contents of the closed loop information table of the C and F stations described above add their own addresses, and (C, D, E, F) become closed loop address elements, and represent the same closed loop. .

Next, a closed loop end point is selected (step S34). Since only one closed loop end point can exist on the same closed loop, in the case of FIG. 14 (A) above, stations C and F find that closed loop end points other than themselves are in the same closed loop from the route information notifications issued by each other. Whether or not it exists is determined by the following procedure. The address registered at the beginning of the address group in the route information notification is the end point station (issuer). When the closed-loop terminal station receives the route information which is not the issuer, the closed-loop end point station refers to the closed-loop information table and determines whether the issuer is in the closed-loop information managed by itself.

In the above case, as shown in FIG. 19, both C and F stations know that the other party (end point) is included in the closed loop information managed by themselves, and the priority end point station is determined. The priority terminal station is determined by first checking whether the end point type of the received route information is a perfect end point, and if it is a perfect end point, the other party is prioritized (the end point type is a perfect end point even though they are on the same closed loop). There is some change in the path and it is judged that the closed loop is broken). In the case of FIG. 14 (A), both C and F stations are “end point type = closed loop”.
In this case, the smaller station address is prioritized as the closed loop end point. That is, in FIG. 14A, the C station has priority, and the C station has F
The route information notification issued by the station is ignored (discarded), and the operation of the closed loop end point is continued.

On the other hand, station F deletes the closed loop information including station C from the closed loop information table. In the case of this example, since there is only one loop registration of the closed loop, there is no registration in the closed loop information table in the F station, the F station ends the operation of the closed loop end point, and shifts to the normal communication station operation.

Also for the case of FIG.
A closed loop end point station is determined among A, B, and F by the same procedure as in (A), and station B with a small station address becomes the closed loop end point. In this case, station A cannot be a closed loop endpoint since it was a full endpoint before it was outside the closed loop. As described above, station B in FIG. 14 (A) and B in FIG. 14 (B)
Both stations detect the closed loop with the route information of FIG.
It can be seen from the contents of the adjacent station table that there are cases where the closed loop end point can be obtained and cases where it is not.

Considering the closed loop end points in the case of FIG. 13A, since this closed loop is a simple closed loop, there is no end point station. Therefore, "a fixed time elapses despite the presence of left and right adjacent stations. But I can't receive the route information notification. "
An event occurs and the station that first detects this event spontaneously transitions to the closed loop endpoint station. Since the detection time of the event that the route information notification of each station cannot be received for a certain period of time is not exactly the same, if any one station first detects and issues the route information notification, the other station shifts to the closed loop endpoint station. Do not take action. In rare cases, a plurality of stations may time out at the same time to make a transition to the closed loop end point station. In that case, as in the example of FIG. 14, the station with the smaller address has priority.

A plurality of entries may exist in the closed loop information table 13 depending on its route configuration. For example, a closed loop as shown in FIG. 21 can be considered. In this case, if stations A and B simultaneously detect that "route information notification cannot be received for a certain period of time" and try to move to the closed loop endpoints together, the route information notifications issued by each of them cause
The contents of the closed loop information table recognized / registered by the stations A and B are as shown in FIGS. 22 (A) and 22 (B).

In the closed loop information table of these two stations, information of closed loops (A, B, C, D) is redundantly registered in both stations, but as described with reference to FIG. By the route information notification issued, it is corrected so that only one closed loop information endpoint exists on the same closed loop (FIG. 22).
(See (C)). Specifically, when the B station receives the route information notification issued by the A station, the address of the A station is smaller than that of the B station. Therefore, the B station searches its own closed loop information table and the closed loop information including the A station. Are all deleted (if there are multiple pieces of closed loop information including station A, all are deleted). In this case, station B deletes the information in the closed loop information table 2 from the closed loop table. However, in this case, since the information of the closed loop 1 still exists in the closed loop information table of the B station, the B station continues to operate the closed loop end point.

The above description is the operation in the case where the stations A and B try to move to the closed loop end point at the same time, but normally, any one station first starts issuing the route information notification. If station A issues it first, the route information notification issued by station A goes around the closed loop of B, C, and E, and B, C, and
Of the E stations, the station that first detects the closed loop of B, C, E moves to the closed loop end point station. In this case as well, the shift to the closed loop end point of a plurality of stations may occur due to the "closed loop simultaneous detection". Since it is the minimum address, station B finally becomes a closed loop end point for the closed loops of B, C, and E, and the contents of the closed loop information table held in station B are the same as above.

When the route configuration includes a closed loop, the determination as to whether the branch station is outside or inside the closed loop is "whether any two of the communication stations adjacent to the same port are included in the route information notification". In the example of FIG. 14, the route configuration is Y branch, that is, the maximum number of adjacent stations in the same port in each branch station is two stations. As shown in FIG. 23 (A), a route having three or more branches can be similarly considered. The adjacent information table in this case is as shown in FIG.

For example, in a C station having a plurality of B, F, and G stations under the control of port 1, the route information notification issued by the perfect endpoint station A goes around the loop, and the route information is A>
Although B>C>D>E>F> (C), the closed loop information D, E, F obtained from this path information includes (B, F,
Since only one station F exists in G), station C is determined to be in the loop. On the other hand, C, F, G under port 2
In the B station having a plurality of stations, the route information obtained by the route information notification issued by the perfect endpoint station A circling the loop is A>B>C>D>E>F> (B). The closed loop information C, D, E, F obtained from the route information includes
Since there are two stations of C and F in (C, F, G), station B is judged to be outside the closed loop.

Next, the operation of the data transmitter / receiver 15 shown in FIG. 2 when each communication station transmits and receives data will be described with reference to the flow charts of FIGS. Referring to FIG. 24, which is a flow chart in the case of data transmission, when there is data to be transmitted (step S41), the learning table 11 is referred to (step S42), and the destination address of the transmission data exists in the learning table 11. (Step S43), data is transmitted only to the relevant port side (step S44), and if the target address does not exist in the learning table, it is broadcast to both ports (step S45).

Further, referring to FIG. 25 which is a flow chart for data reception, even when data is received from an arbitrary port (step S51), when the received data packet is not addressed to itself (step S52). After referring to the learning table (step S53) and confirming that the destination address of the received data is not on the input port side (step S54), the data is transferred to the opposite port (step S55). If there is a data destination address on the input port side, the corresponding received data is data that does not need to be transferred by the local station and is therefore not transferred but discarded (step S57). The data addressed to itself is terminated (step S56).

[0049]

INDUSTRIAL APPLICABILITY The present invention can be issued by an end point station even if the communication path configuration dynamically changes in an environment in which all nodes on the network represented by a point-to-point digital communication network operate as bridges. The learning table held by each station is quickly updated by the route information. In addition, even when the communication path includes a closed loop configuration, the logical path separation by the closed loop end point has an effect of easily preventing a data storm of broadcast data.

[Brief description of drawings]

FIG. 1 is a schematic diagram of a network configuration to which the present invention is applied.

FIG. 2 is a schematic configuration diagram of an embodiment of a communication station of the present invention.

FIG. 3 is a diagram showing a learning table.

FIG. 4 is a diagram showing an adjacent station information table.

FIG. 5 is a diagram showing a closed loop information table.

FIG. 6 is a flow chart showing an endpoint station determination process of each communication station.

7A is a health check request message issued by each communication station, and FIG. 7B is a response message thereof.

8A is an explanatory diagram of an end point station in a linear network topology, and FIG. 8B is a diagram showing the contents of an adjacent station information table of each communication station in this case.

9A is an explanatory diagram of an end point station in a network topology including branches, and FIG. 9B is a diagram showing the contents of an adjacent station information table of each communication station in this case.

FIG. 10 is a flowchart showing a route information notification transmission process of an end point station.

FIG. 11 is a flowchart showing a route information notification receiving process of each communication station.

FIG. 12A is a diagram showing a route information notification message;
(B) is a figure which shows the example.

FIG. 13 is a diagram showing a network topology including a closed loop.

FIG. 14 is a diagram for explaining a closed loop end point station in a network topology including a closed loop.

15 is a diagram showing the contents of an adjacent information table of each communication station in the network topology of FIG.

FIG. 16 is a flowchart showing an outline of a procedure for determining a closed loop end point station in a network topology including a closed loop.

17 is a diagram showing an example of route information notification in the network topology of FIG.

18 is a diagram showing an example of route information notification in the network topology of FIG.

19 is a diagram showing an example of route information notification in the network topology of FIG.

20 is a diagram showing an example of route information notification in the network topology of FIG.

FIG. 21 is a diagram showing a network topology including multiple closed loops.

22 is a diagram showing the contents of a closed loop information table in the network topology of FIG.

FIG. 23A is a diagram showing a network topology when the number of branches is three. (B) is a diagram showing the contents of the adjacent station information table at the branch station in that case.

FIG. 24 is a processing flow at the time of data transmission in each communication station.

FIG. 25 is a processing flow when data is received by each communication station.

[Explanation of symbols]

1, 2 ports 10 communication stations 11 Learning table 12 Adjacent station information table 13 Closed loop information table 14 Control unit 15 Data transmitter / receiver

Claims (8)

(57) [Claims] 1. A communication network system comprising a plurality of communication stations that perform point-to-point communication with both adjacent stations and make a routing decision of communication data by referring to a learning table, each of the communication stations , Means for exchanging a request message for confirming connection and its response message between adjacent stations, means for judging whether or not the own station is an endpoint station according to the presence or absence of the response message, and the endpoint If it is determined to be a station, it issues a route information notification indicating the communication route with the end point station as its own address and broadcasts it, and if it is determined to be another station, receives the route information. see contains means for inserting a local address in the notification broadcasts propagated to the next station, and means for updating the contents of the learning table in accordance with the route information notification received, the Communication station determines a point station of the route information notification
In addition to occurrence, it controls the end and the end of all broadcast data.
A communication network system characterized by this . 2. A communication network system according to claim 1, wherein transfer of said request message and its response message, characterized in that the manner performed in a constant cycle. 3. A communication network system according to claim 1 or 2, wherein said issuing of the route information notification is characterized in that the manner performed in a constant cycle. 4. The communication station further comprises means for determining whether or not the own station exists in the closed loop by referring to the route information notification, and means for judging if the own station exists in the closed loop. Means for determining whether or not the station should be a closed-loop terminal station, and when it is determined to be the closed-loop terminal station, in addition to the generation of the route information notification, manages its end and the end of all broadcast data. The communication network system according to any one of claims 1 to 3 , characterized in that. 5. A communication control method in a communication network system configured by a plurality of communication stations that perform point-to-point communication with both adjacent stations and make a routing decision of communication data by referring to a learning table, At each station, the step of sending and receiving a request message and its response message for confirming the connection between adjacent stations, and the step of judging whether or not the own station is an endpoint station according to the presence or absence of the response message of If it is determined to be the end point station, it issues a route information notification indicating the communication path with the end point station as its own address and broadcasts it, and if it is determined to be another station, receives it. The step of inserting the own station address into the route information notification and broadcast-transmitting to the next station, and the learning table according to the received route information notification. The viewing including the step of updating the content, communication station determines that the endpoint stations of the route information notification
In addition to occurrence, it controls the end and the end of all broadcast data.
Communication control method characterized in that had Unishi. 6. The communication control method according to claim 5, wherein the request message and the response message thereof are transmitted and received at a constant cycle. 7. The communication control method according to claim 5, wherein the issuance of the route information notification is performed in a constant cycle. 8. Further, in each of the communication stations, a step of judging whether or not the own station exists in a closed loop by referring to the route information notification, and a case of judging that the own station exists in the closed loop, Includes the step of determining whether or not the own station should be a closed loop end point station, and when it is determined to be the closed loop end point station, in addition to the generation of the route information notification, its end and the end of all broadcast data. The communication control method according to any one of claims 5 to 7, wherein the communication control method is controlled.