JPH0795736B2 - Loop network polling method - Google Patents

Description

TECHNICAL FIELD The present invention relates to a polling method for a loop network.

(Prior art and its problems) Various types of communication such as real-time communication such as voice (hereinafter referred to as synchronous communication), standby communication such as data (hereinafter referred to as asynchronous communication), etc. Research and development aimed at accommodating in the same loop network is in progress, and one of them is a paper (No. SE83-107) published by the Exchange Research Group of the Institute of Electronics and Communication Engineers in 1983 by Shimizu. There is a loop network described in "Study" (Reference 1). The outline of the loop network described in Reference 1 will be briefly described. As shown in FIG. 1, the loop network is composed of node stations 1, 2, 3, 4, 5 and 6 for accommodating various terminals and computers, and a trigger station 7 for polling these node stations. They are connected by (hereinafter referred to as a loop) 8. As shown in FIG. 2, the trigger station 7 sends a trigger signal TRG at a constant cycle T to give a time origin to each node station, and polls the communication of the highest priority level. When the polling of the highest priority level is completed, the polling signal POL for polling the communication of the next highest level is sent. Thereafter, polling is repeated in the order of higher priority level within the cycle T, and when the cycle T elapses, the polling for the communication of the highest priority level is returned to. Each node station sends a packet PKT in response to these priority polling. Since the communication with the highest priority level is polled prior to other communication in every cycle T, by giving the highest priority level to the synchronous communication, it is possible to provide the real-time service for the synchronous communication. Further, by giving priority to asynchronous communication as well, various services such as call setting communication and high / low priority data communication can be dealt with. In a communication system in which the priority of communication polled at regular intervals returns to the highest priority level, such as in such a network, unfairness in transmission opportunity occurs depending on the physical position of each node station, as communication at a lower priority level increases. . That is, in the first loop network, the node station 1 closest to the trigger station 7 has the largest transmission opportunity and the farthest node station 6 has the smallest transmission opportunity. To deal with this,
In the conventional example, as shown in FIG. 3, various communications are given a main priority in accordance with a communication form or a service request, and an asynchronous communication in which transmission opportunity becomes unfair, that is, a communication having a main priority of 3 or less is performed. On the other hand, two levels of subordinate priority are given. (Synchronous communication uses the call setting communication as described in detail in this conventional example, so that the transmission opportunity within the cycle T is guaranteed, so it is not necessary to give the sub priority.) In Fig. 3, from priority levels 7 to 1 which are the same as the primary priority and the secondary priority (the lower the number, the lower the priority).
Is giving.

Next, the fairness of the transmission opportunity and its problems in the conventional example will be described with reference to FIGS. 4 (a-1) and (a-2). FIG. 4 (a-1) shows a case where the node station 4 cannot transmit in a certain cycle. It is assumed that a transmission request of main priority 2 occurs at the node station ₄ at time R ₄ . At the time of occurrence, the subordinate priority is given to the lower one of the two levels. That is, priority level 3 is given. The trigger station 7 sequentially polls the priority levels 7, 6, ..., 3 by the trigger signal TRG and the polling signals POL6 ,. Node stations 1, 3 upstream of node station 4 based on polling signal POL3
Sends packets PKT1 and PKT3 respectively, and then the cycle T
It is assumed that transmission of all node stations is prohibited due to the progress of. Since the node station 4 could not transmit even though the polling priority level became the same as the priority level 3 of its own transmission request, the node station 4 raises the slave priority level to 2 and sets it to the priority level 4 in the next cycle. Fig. 4 (a-
2) shows the signal on the loop 8 in the next cycle. At time R ₂ of this cycle, it is assumed that a transmission request of main priority 2 is issued to the node station 2 located upstream of the node station 4. In this case, since the transmission request is given priority level 3 as described above, the node station 4
Sends a packet PKT4 based on the polling signal POL4, and the node station 2 sends a packet PKT2 based on the polling signal POL3. That is, the node station 4 can transmit before the node station 2, and the transmission opportunities are faired. However, in this case, as shown in FIG. 4 (a-2), there is no node station that transmits based on the polling signal POL6. This is because the polling of priority level 5 is completed in the cycle before this cycle, that is, the cycle shown in (a-1) of FIG. This is because there is no transmission request of the third priority, the secondary priority of 2, that is, the priority level 6. Therefore, the priority level 6 polling has a dead time Tm. This time Tm is equal to the propagation time td of one loop, but the higher the transmission speed of the loop 8, the larger the loss of transmission capacity due to the dead time Tm (loop speed × dead time Tm).

Regarding the method of giving priority by the polling signal, there is one described in Patent Application No. 58-124481 "Priority recognizing method in loop type network". Even if it does, the same problem occurs.

(Object of the Invention) An object of the present invention is to prevent the above dead time.

(Structure of the Invention) According to the present invention, a plurality of node stations and one trigger station for polling these node stations are connected by a loop-shaped transmission line, and the trigger station polls from a communication having a high priority within a certain period. However, in the polling method of the loop network that gives the main priority to each communication and the two-level sub-priority to the main priority, the node station has two levels when a transmission request for communication of a certain main priority occurs. If the transmission control is activated at the lower level of the sub-priority, and if transmission is not possible within the fixed period, the transmission control is activated at the high level of the sub-priority in the next period, and the trigger station The priority polled in each cycle is monitored, and the polling of the low level sub-priority of 2 levels is completed in a certain cycle. For the main priority higher than the main priority, only the polling for setting the sub-priority to the low level is performed in the next cycle, and the polling for the low-level sub-priority is not completed in the certain cycle. A polling method for a loop network is obtained which is characterized in that the secondary priority performs high level and low level polling in the next cycle with respect to the main priority.

(Embodiment) FIG. 5 shows an example of a signal frame used in the first embodiment of the present invention. The signal frame in FIG. 5 is the preamble PR,
Start delimiter SD, trigger signal TRG bit
It consists of TG, priority level field PI, destination address DA, source address SA, information field INO and end delimiter ED.

The polling operation in the first embodiment is shown in FIG.
This will be described with reference to 1) and FIG. Assume that polling is completed at the priority level 3 in a certain cycle as shown in FIG. 4 (a-1) and the next cycle is started. When the trigger station 7 recognizes that the final polling has reached the priority level 3, in the next cycle, the trigger station 7 only polls the sub-priority 1 for the main priority higher than the above-mentioned level. To In this case, priority level 3
The secondary priority 2 is not polled for the primary priority 3, which is higher than the primary priority 2. That is, as shown in FIG. 6, after the polling of the priority level 7 by the trigger signal TRG,
For the priorities less than the main priority 2 for which the priority level 5 is polled, the secondary priority is 2.1, that is, the priority level 4.3 ,. Therefore, since the polling of the priority level 6 is not performed, there is no useless polling as in the conventional example. In addition, fairness of the transmission opportunity of each node station is not impaired.

Next, FIG. 7 shows the structure of the trigger station 7 used in this embodiment. The trigger station 7 in FIG. 7 is an input terminal.
It is connected to the loop 8 by 10 and the output terminal 11. When the control unit 16 is notified of the elapse of the period T by the timer 19, the signal frame (see FIG. 5) in the transmission buffer 17 is bit-transmitted.
Set TG = 1, PI = 111 (7) and switch 13 to send buffer 17
Connect to the side and send the trigger signal TRG. This trigger signal T
RG is input to the buffer 12 after one round of the loop 8 and supplied to the signal detection circuit 14. The signal detection circuit 14 supplies the value of the PI field to the register 18. After receiving the trigger signal TRG, the control unit 16 returns the switch 13 to the buffer 12 side to allow the packet transmitted from each node station to pass through and to perform transmission right acquisition control. When the transmission right is acquired, the transmission frame in the transmission buffer 17 is set to the bit TG = 0, PI is set to the above-mentioned value, and the polling signal is transmitted. For example, if the priority level of final polling in the previous cycle is 4 or less, PI = 101 (5), and if the priority level of final polling is 6 or 5, PI = 110 (6). Control unit 1
6 repeats the above polling control until the timer 19 passes the elapse of the period T. Each time the polling signal goes through the loop 8 and is supplied to the signal detection circuit 14, the value of PI is supplied to the register 18 and updated. Therefore, when the period T has elapsed, the priority level of the final polling of the period is stored in the register 18. The control unit 16 determines the priority level to be polled in the next cycle based on the value of the register 18. That is, if the value stored in the register 18 after the elapse of the period T is a binary value 110 (6) or 101 (5), polling of all priority levels is performed, and 100 (4) or 011 ( If 3), the polling of the main priority 3 is completed, so the priority level 6 is skipped and the polling is performed, and 010 (2) or 001
If (1), polling with main priority levels of 3 and 2 has been completed, so skip priority levels 6 and 4 and perform polling.
If it is (0), polling of the main priority levels 3, 2, and 1 has been completed, so the priority levels 6, 4, and 2 are skipped and polling is performed. Register 18 is the timer at the beginning of the next cycle.
It is reset and initialized by 19. Note that the transmission right acquisition control in the present embodiment is performed based on the preamble PR and the end delimiter ED detected by the signal detection circuit 14, but the transmission right acquisition method does not limit the present invention,
The method in the conventional example described above may be used, and the description thereof is omitted in this specification. Further, the signal frame of the loop 8 is supplied to the receiving unit (not shown) via the receiving buffer 15.

A node station used in this embodiment is shown in FIG.
The node station of FIG. 8 has an input terminal 20 and an output terminal.
It is connected to the loop 8 by 21 and includes a buffer 22, a switch 23,
It is composed of a signal detection circuit 24, a reception buffer 25, a control unit 26, a transmission buffer 27, and a timer 29. The signal detection circuit 24 determines the value of the priority field PI of the passing signal frame by the control unit 26.
And the trigger bit TG = 1 is detected, the timer 29 is reset. The control unit 26 compares the priority level of the transmission request with the PI from the signal detection circuit 24, and if the former is the same as or higher than the latter, starts the transmission right acquisition control. When the transmission right is acquired, the PI given by the signal detection circuit 24
To the signal frame in the transmit buffer 27 and switch 23
And to the loop 8 via the output terminal 21. When the timer 29 notifies the elapse of the period T while the transmission right cannot be acquired, the control unit 26 controls the priority level of the transmission request as follows. When the secondary priority of the transmission request is 1 and the priority level given by the polling signal is equal to the priority level of the transmission request, the secondary priority is set to 2 and the priority level is increased by 1. In other cases, the priority level remains unchanged. Note that this node station has the same function as the node station in the above-mentioned conventional example, and there is no new function added for implementing the present invention.

Next, a second embodiment of the present invention will be described. FIG. 9 shows the structure of a signal frame used in this embodiment. In the configuration of FIG. 9, a bit PL indicating whether or not it is a polling signal is provided instead of the priority level field PI. The polling method in this embodiment will be described with reference to FIGS. 10 (a-1) and (a-2). At the beginning of the cycle T, the trigger station 7 sends the trigger signal TRG with the bit TG = PL = 1 and the polling signal POL with TG = 0 and PL = 1 within the cycle T. Each node station counts the number of times the polling signal POL has passed within the period T, and recognizes the priority level of the polled communication correspondingly. First
In Figure 10 (a-1), each node station has a trigger signal.
TRG recognizes priority level 7 and next polling signal POL recognizes priority level 6. Each time the polling signal POL passes, the priority level decreases by one.

When the trigger station 7 sends the polling signal POL three times as shown in FIG. 10 (a-1) and the cycle T ends, the priority level given by the last polling signal POL in this cycle is 4 Recognize. In this case, in the next cycle, it is necessary to perform the polling of the priority level 5 and the polling of the priority level 5 instead of the polling of the priority level 6 as shown in the first embodiment. This method is shown in FIG. That is, after the polling of the priority level 7, two polling signals POL are continuously transmitted. The previous polling signal POL of the two gives a priority level of 6, but the subsequent polling signal POL gives a priority level of 5. By the above control, it is possible to carry out polling without waste as in the first embodiment.

FIG. 11 shows the structure of the trigger station using the second embodiment. The configuration of FIG. 11 has a transmission buffer 37 and a counter.
The structure is the same as that of FIG. 7 except 38. The counter 38 is initialized to a value of 8 every period T by the timer 19 and is decremented every time the signal detection circuit 14 detects the bit PL = 1. When the counter 38 reaches the count value 0, it stops counting until it is initialized again. The control unit 16 is a counter
The priority level is recognized from the output value of 38, and if there is a priority level that does not perform polling, the control output 161 causes the transmission buffer 37 to send out two polling signals.
If there is no priority level at which polling is not performed, control is performed so that only one polling signal is transmitted.

FIG. 12 shows the configuration of the node station used in this embodiment. In the node station of FIG. 12, the counter 48 has the bit TG = 1 detected by the signal detection circuit 24.
Is initialized by and is decremented by bit PL = 1. The control unit 26 recognizes the priority level polled by the output value of the counter 48 and performs the same control as in the first embodiment.

It should be noted that the trigger stations shown in FIGS. 7 and 11 monitor the signals on the loop and recognize the priority level of the communication to be polled. However, the triggers to be sent out as shown in FIGS. 13 and 14 are sent. The priority level may be recognized by monitoring the signal and the polling signal.

The trigger station shown in FIG. 13 is used in the first embodiment, and the register 18 stores the value of PI set by the control unit 16. The trigger station shown in FIG. 14 is used in the second embodiment, and the counter 38 counts the number of trigger signals and polling signals sent from the transmission buffer 37.

(Effects of the Invention) As described above, according to the present invention, wasteful polling is not performed without impairing the fairness of the transmission opportunity of each node station, so that the transmission efficiency of the loop transmission line can be improved. Further, in order to carry out the present invention, it suffices to give the above-mentioned function only to the trigger station, and no new function is required for the node station. Therefore, the present invention can be economically carried out with respect to the existing conventional system. Can be done.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a loop network, and FIG.
Figures 4, (a-1), (a-2), 6 and 10 (a-1), (a-2) show the flow of signals on the loop, and FIG. 3 shows Diagram showing priority level assignment, Fig. 5,
FIG. 9 is a diagram showing the structure of a signal frame, FIG. 7, and FIG.
FIGS. 13, 13 and 14 are block diagrams showing the configuration of the trigger station, and FIGS. 8 and 12 are block diagrams showing the configuration of the node station. In the figure, 1,2,3,4,5,6 are node stations, 8 is a loop, 12,2215,25,17,27,37 are buffers, 14 and 24 are signal detection circuits, and 16 and 26 are control units. 18 is a register, 19 and 29 are timers, 38 and 48 are counters, and 13 and 23 are switches.

Claims (1)

[Claims] 1. A plurality of node stations and one trigger station for polling these node stations are connected by a loop-shaped transmission line, and the trigger station polls from a communication with a high priority within a certain period and for each communication. In the polling method of the loop network that gives the main priority and the sub-priority of two levels to the main priority, the node station selects the lower level of the two levels when a transmission request for communication of a certain main priority occurs. If the transmission control is activated at the sub-priority level, and if transmission is not possible within the fixed period, the transmission control is activated at the high level sub-priority level in the next period, and the trigger station is polled at each period. The priority is monitored, and in a certain cycle, the polling of the sub-priority of the lower level of the two levels is completed In the next cycle, only the polling that sets the sub-priority to the low level is performed for the main priority, and the polling of the above-mentioned low-level sub-priority is not completed in the above-mentioned certain cycle. In the next cycle, the polling method of the loop network is characterized in that high-level and low-level polling is performed in the secondary priority.