JPH04360259A - Transmission control system by polling - Google Patents

Abstract

PURPOSE:To rapidly and efficiently execute data transmission from a slave station to a main station by a polling system. CONSTITUTION:At the time of receiving (SC1 to SC3) a polling message which is not transmitted to the slave station, a data transmission control part in the slave station monitors (SC4 to SC5) whether data are sent from another slave station to a bus during an optical time generated by random number generation, and when no data are transmitted, transmits (SC6) a response message including transmission data to the master station.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention uses a polling method for data transmission between the master station and each slave station in a system in which a plurality of slave stations are connected to one master station via a common transmission path (bus). This paper relates to a transmission control method using polling.

0002

[Prior Art] When multiple terminal devices are connected to one control device (host computer, etc.) via a common transmission path such as a bus using a multi-drop method, the connection between the control device and each terminal device is A polling method is usually used for data transfer control.

[0003] In this polling method, one main station (or control station) sends information to multiple slave stations (or dependent stations).
By sequentially transmitting polling messages inviting message transmission (performing polling) and giving message transmission rights only to the slave stations specified in the polling messages, data sent from multiple slave stations to the master station can be , a transmission control procedure (transmission protocol) that prevents collisions on a common transmission path such as a bus and ensures data transfer between a master station and a slave station. In this polling method, the polled slave station transmits data to the master station if it has data to send.
If there is no data to be sent, a message to that effect is returned or no response is made (no response is made).

On the other hand, in addition to the above-mentioned polling method, the master station polls all slave stations simultaneously, and the slave stations that have data to send respond to the polling by transmitting data to the master station. There is also a polling method. In this polling method, in order to avoid data collisions occurring on a common transmission path due to simultaneous data transmission from multiple slave stations, the slave stations transmit data before transmitting data.
After a delay of an appropriate amount of time due to random number generation, it is checked whether or not another station is transmitting data, and if it is, it performs controls such as waiting for data transmission.

[0005]

However, the former method has the disadvantage that as the number of slave stations increases, the cycle at which each slave station receives polling from the master station (polling waiting time) becomes longer. Moreover, even when the amount of data transmitted between the master station and the slave station is small, there is a drawback that the waiting time for polling at the slave station increases.

Furthermore, in the latter method, even if the number of slave stations increases, the polling period of each slave station does not change. However, as the number of data transmitted from slave stations increases, the probability that data will collide on the common transmission path increases. Therefore, depending on factors such as the location on the common transmission path to which each slave station is connected and differences in performance between each slave station, a situation may occur where a particular slave station is almost unable to obtain transmission rights (no longer able to transmit data). There was a drawback.

The disadvantage of the former method is thought to be due to the fact that only polled slave stations are given the right to transmit data to the master station. In addition, the disadvantage of the latter method is that if there is a large amount of data to be sent from the slave stations to the master station, even though there is a high probability that multiple slave stations will transmit data to the master station at the same time, all slave stations will This is thought to occur because polling is performed simultaneously for both.

[0008] Then, the polling for the slave station is as follows.
Each station is polled in turn, and if the polled slave station has no data to send to the master station, the unpolled slave station that has data to send to the master station will wait until the next polling time. If the data can be sent to the master station by using this function, it can be used not only when the amount of data exchanged between the master station and the slave stations is small and the number of slave stations is large, but also when the amount of data exchanged between the master station and slave stations is large. It is clear that even when a large amount of data is exchanged, each slave station can transmit data to the master station more quickly and efficiently than before.

The problem of the present invention is that polling of the slave stations is performed one by one in order, and if the polled slave station does not have data to send to the master station, polling is performed in which there is data to send to the master station. The subordinate who has not been
The purpose is to make it possible to send data to the main station using free time until the next polling.

0010

Means for Solving the Problems The means of the present invention are as follows. The present invention includes one main station 1 (see the functional block diagram in FIG. 1; the same applies hereinafter) and a plurality of slave stations 2 (2-1, 2-2, . . . ) that transmit data to the main station 1. 2-N) are connected to one common transmission path 3,
The master station 1 polls each slave station 2-1, 2-2, ... 2-N one by one, and polls each slave station 2-1, 2-2, ... 2-N.
A transmission control method based on polling is assumed, in which N is invited to transmit data. And each slave station 2-1,
2-2,...2-N have the following means 2a, 2b. The time generation means 2b is a means for arbitrarily (for example, randomly generating) a time within a predetermined range. For example, during the time within the above predetermined range, the slave station 2-i (i=1, 2,...N) polled by the master station 1 sends a message (for example, a message consisting of one character) indicating that there is no data to be transmitted. )
After the master station 1 polls any slave station 2-i for a period sufficiently longer than the time required to send the slave station 2-i,
A time shorter than the maximum time to wait for a response message from (
(equal to the time until the next polling). The monitoring means 2a monitors whether data is being sent out on the common transmission path 3. Each slave station 2-1, 2-2, . . . 2-N having the means 2a, 2b has the following characteristics. In other words, each polled slave station 2-1, 2-2, ... 2-N, if there is data to be transmitted to the master station 1, immediately transmits the data to the master station via the common transmission path 3. Each slave station 2-1, 2-2, . During the period, the monitoring means 2a monitors the data on the common transmission path 3, and when it is determined that no data is sent from another slave station 2 to the common transmission path 3 during the above-mentioned arbitrary time, the data to be transmitted is is transmitted to the main station 1 via the common transmission path 3.

[0011]

[Operation] The operation of the means of the present invention is as follows. Main station 1
If the slave station 2 polled by the master station 2 has transmission data for the master station 2, it immediately transmits a response message containing the transmission data to the master station 1 via the common transmission path 3; If there is no such message, send a message to that effect to the main station 1 via the common transmission path 3, or
Or, it does not respond to the main station 1. On the other hand, when the slave station 2 that has not been polled has data to be transmitted to the master station 1, the time generating means 2b generates an arbitrary time within the above-mentioned predetermined range, and during that arbitrary time, the monitoring means 2
a, it continues to monitor whether there is data sent from other slave stations 2 on the common transmission path 3, and during that arbitrary time, data is sent from other slave stations 2 on the common transmission path 3. If not, a response message containing the above transmission data is transmitted to the main station 1 via the common transmission path 3. Therefore, polling of slave stations is performed one by one in order, and if a polled slave station has no data to send to the master station, the slave station that has not been polled but has data to send to the master station will be the next one. Data can be sent to the main station using the free time before polling.

0012

[Embodiment] An embodiment will be described below with reference to FIGS. 2 to 11. FIG. 2 is a system configuration diagram of a computer network.

As shown in the figure, this computer network consists of a main station 20 consisting of a host computer, etc., and four slave stations 30 (30-1, 30-1, 30-1,
-2, 30-3, 30-4) are connected to a bus 50, which is a common transmission path, by a connection cable 40.

Next, the internal configuration of the main station 20 is shown in FIG. As shown in the figure, the main station 20 includes a main station control section 21, a data reception control section 22, a polling list memory 23,
and line receiver 2, which is a three-state buffer.
5 and a line driver 26, each dependent station 3
0-1, 30-2, 30-3, and 30-4, and each dependent station 30-1 to 30-4 receives data transmitted in response to the polling.

The line receiver 25 has its input terminal connected to the connection cable 40, and its control terminal and output terminal both connected to the data reception control section 22. Also,
The line driver 26 has an output terminal connected to the connection cable 4.
0, and both the control terminal and the input terminal are connected to the data reception control section 22.

[0016] The data reception control unit 22 is connected to the main station control unit 21.
receives the transmission data of each slave station 30-i (i=1 to 4) output on the bus 50 via the line receiver 25, and receives the transmission data of each slave station 30-1 to 30 through the line receiver 25.
-4 is output on bus 50 via line driver 26.

The master station control section 21 controls the data reception control section 22 as described above, and also processes the data transmitted from each slave station 30-i. Also,
The polling list memory 23 consists of, for example, ROM (read only memory), RAM (random access memory), etc., and is used to poll each slave station 30-i. slave 30
-1 to 40-4 terminal addresses are stored.

Next, in FIG. 4, slave stations 30 (30-1 to 30-
4) shows the internal configuration. As shown in the figure, the slave station 30 includes a slave station control section 31, a data transmission control section 32, a terminal address memory 33, and a line receiver 35 and a line driver 36 consisting of a three-state buffer.

The line receiver 35 has an input terminal connected to the connection cable 40, a control terminal and an output terminal both connected to the data transmission control unit 32, and the line driver 3
6 has an output terminal connected to the connection cable 40, and both a control terminal and an input terminal connected to the data transmission control section 32.

[0020] The data transmission control section 32 is connected to the slave station control section 31.
It operates under the control of the main station 20 via the line receiver 35, receives the polling message sent from the main station 20, detects polling for its own station, and transmits the transmission data to the main station 30 output by the slave station control section 31. , and is sent onto the bus 50 via the line driver 36 and the connection cable 40 at a predetermined timing, as will be explained in detail later.

The slave station control section 31 controls the data transmission control section 32 as described above, and also controls the timing of transmitting data to the master station 30, as will be explained in detail later.

The terminal address memory 32 is, for example, a ROM.
, RAM, etc., and stores the terminal address of its own station. Next, the operation of the data reception control section 22 of the main station is shown in FIG.
This will be explained with reference to the flowchart.

The data reception control unit 22 first extracts the terminal address from the head of the polling list stored in the polling list memory 23, and stores the terminal address in the ADR register (SA1).

[0024] Next, polling list update processing SA2, which will be explained in detail later, is performed, and the first terminal address of the polling list set in the ADR register is moved to the end of the polling list, and the second and subsequent terminal addresses are sequentially updated. , move up one by one. Through this polling list update process SA2, the terminal address to be polled (that is, the slave station 30 to be polled) is cyclically changed.

Next, a polling message for polling the slave station 30-i having the terminal address stored in the ADR register is output onto the bus 50 via the line driver 26 (SA3).

FIG. 6 shows an example of the structure of the polling message. In the example shown in the figure, the polling message is STX (Star) indicating the start of the text from the beginning.
t of Text), an identifier P indicating that it is a polling message, a terminal address ADR, an ETX (End of Text) indicating the end of the text, and a horizontal parity bit for detecting bit errors in the polling message. BCC (Block C)
heckCharacter).

The flowchart in FIG. 5 will be explained again. In the above process SA3, after sending out the polling message, any slave station 30 via the line receiver 25
-i (i=1 to 4) enters a waiting state to receive a response message transmitted from the terminal (SA4).

[0028] If no response message is received from any slave station 30-i within the predetermined time to, it is determined that a timeout error has occurred (SA5), and the above process SA1 is performed again.
~SA3 is performed, and the process moves to polling of the next slave station 30 (the slave station 30 having the terminal address at the head of the polling list).

On the other hand, if a response message is received from any slave station 30-i (i=1 to 4) before the timeout error occurs, it is determined whether the data of the received response message is normal or not. (SA6)
, if a data error has occurred in the response message, the next slave station 30 is polled (SA1 to SA3) in the same way as for the timeout error described above.

On the other hand, if a normal response message is received before a timeout error occurs, the line driver 26
An acknowledgment (ACK) is returned via (SA7).

[0031] Here, an example of the structure of the above response message is shown in FIG. The response message shown in the figure includes, from the beginning, the STX, the identifier R indicating that it is a response message, the terminal address ADR of the slave station 30 that sent this response message, and the data section in which the data sent to the master station 20 is stored. , the ETX, and the BCC.

Although this is a feature of this embodiment, the terminal address in the response message received in the above process SA4 does not necessarily correspond to the slave station 3 polled in the above process SA3.
The terminal address is not limited to 0, but may be the terminal address of a slave station that has not been polled. The response message in this case is processed as a normal response message. In other words, as will be explained in detail later,
A slave station 30 that has not been polled may also be able to send a response message to the master station 20.

Returning to the flowchart of FIG. 5 again, the data reception control unit 22 extracts the terminal address from the received response message, stores the terminal address in the ADR register (SA8), and then , performs a process similar to the polling list update process SA2 described above (SA9).

By this polling list update process SA9, the terminal address of the slave station 30 that sent the response message is moved to the end of the polling list. As described above, in this embodiment, there is a possibility that a slave station 30-j (i≠j: j=1 to 4) other than the polled slave station 30-i transmits a response message. By performing polling list update processing SA9,
Slave station 3 that sent the response message to master station 20
The next polling of 0-i (i=1 to 4) will always be performed after polling of all slave stations 30 other than that slave station 30-i has been completed.

After performing the polling list update process SA9, the data reception control section 22 extracts the data section from the received response message and outputs the data section as received data to the main station control section 21 (SA10). )
.

[0036] Accordingly, the main station control section 21 performs the relevant processing on the received data. Next, main station 20
The polling list update processes SA2 and SA9 performed by the data reception control section 22 will be explained with reference to the flowchart of FIG.

In these processes SA2 and SA9, the data reception control section 22 first sets the starting address of the polling list in the working pointer P (hereinafter simply expressed as pointer P) (SB1).

Next, it is determined whether the terminal address stored at the address indicated by pointer P in the polling list and the terminal address stored in the ADR register are equal (SB2), and if they are not equal, pointer P is After adding "1" to (SB3), the above process SB2 is repeated.

Therefore, the above process SB2 is repeated until the terminal address stored at the address indicated by the pointer P of the polling list and the terminal address stored in the ADR register become equal. When the above two terminal addresses become equal, it is determined whether the pointer P is equal to the final address of the polling list (SB4), and if not, the address one address after the address indicated by the pointer P of the polling list is determined. After moving the terminal address stored in , to the address indicated by pointer P on the polling list, "1" is added to pointer P (SB5).

This process SB5 is repeated until it is determined in the process SB4 that the address indicated by the pointer P is equal to the final address of the polling list. If it is determined in the above process SB4 that the address indicated by the pointer P is equal to the final address of the polling list, the terminal address stored in the ADR register is stored in the final address indicated by the pointer P of the polling list (SB6). .

Here, in order to facilitate understanding of the operations of the polling list update processes SA2 and SA9 described above, FIG.
An example of updating the polling list performed by the above processes SA2 and SA9 will be described with reference to the examples shown in FIGS. 10(a) and 10(b).

First, the polling list 100 is created as shown in FIG.
As shown in a), from the beginning, the address is ADR1.
(first address), ADR2, ADR3, ADR4 (last address) are assigned, and each address ADR1,
ADR2, ADR3, and ADR4 each have slave stations 30
The following explanation will be given assuming that terminal addresses "1", "2", "3", and "4" of -1, 30-2, 30-3, and 30-4 are stored.

In such a case, the data reception control unit 22 first selects the slave station 30- with the terminal address "1" stored in the first address ADR1 of the polling list.
1 is polled, and the terminal address "1" is stored in the ADR register (SA1 in the flowchart shown in FIG. 5).

[0044] Next, the data reception control section 22 performs the polling list update process SA2.
2, the start address ADR1 of the polling list is first set in the pointer P (SB1 in the flowchart shown in FIG. 8), and in the subsequent process SB2, the terminal address "1" stored in the address "ADR1" indicated by the pointer P is immediately set. ” and the terminal address “1” stored in the ADR register are determined to be equal. and,
Thereafter, the pointer P is sequentially updated to "ADR2" and "ADR3" while the address "ADR" of the polling list 100 is updated.
As shown in FIG. 9(b), the terminal addresses "2", "3", and "4" stored in "DR2", "ADR3", and "ADR4" respectively correspond to the addresses "ADR1" and "ADR1" of the polling list 100, respectively. It is moved to "ADR2" and "ADR3" (processing SB5 is repeated three times). Then, after pointer P is updated to address "ADR4", ADR
The terminal address "1" stored in the register is stored in the last address "ADR4" of the polling list 100, as shown in FIG.
→SB6).

After updating the contents of the polling list 100 as shown in FIG. A polling message is transmitted to the slave station 30-1 having the terminal (SA3).

[0046] Then, in the following process SA4, the slave station 30-4
Upon receiving a normal response message from the terminal, the data reception processing section 22 sets the terminal address "4" of the slave station 30-4 stored in the received response message in the ADR register in the subsequent processing SA8. , then performs polling list update processing SA9.

The contents of the polling list 100 and the ADR register immediately before this process SA9 is performed are shown in FIG.
It is as shown in a) (equivalent to FIG. 9(b)). Therefore, unlike the process SA2 described above, in this process SA9, process SB3 is repeated twice, and in the second process SB3, the address "ADR3" is set to the pointer P.
After is set, the process moves from process SB2 to process SB4.

After determining "NO" in the process SB4, the process SB5 is performed and the polling list 10 is
The terminal address "1" stored in the address "ADR4" of 0 (see figure (a)) is transferred to the address "ADR3".
After performing the process of moving the pointer to P (see (b) in the same figure),
The final address of polling list 100 is "ADR4"
Change to Then, "YES" is determined in the above process SB4, and then the above process SB6 is performed, and the terminal address "4" stored in the ADR register is stored in the last address "ADR4" of the polling list 100.

As a result, the terminal address "4" of the slave station 30-4 that sent the received response message is stored in the last address "ADR4" of the polling list 100. Then, by the next polling process SA3, the first address "ADR1" of the polling list 100 is
slave station 3 with terminal address "2" stored in "
Poll 0-2.

In this way, the polling list update process S
In response to the polling, A2 moves the terminal address of the slave station 30 at the head of the polling list that performed the polling to the end of the polling list, and moves the other terminal addresses on the polling list one by one upward in the polling list. The polling list processing SA9 is
Actual received response message (polled slave 3)
The terminal address of the slave station 30 that sent the response message (not necessarily the response message from 0) is moved to the end of the polling list, and the terminal address connected after the terminal address moved to the end of the polling list is moved to 1. This is the process of moving them up the polling list one by one.

[0051] Next, each slave station 30 (30-1 to 40-4
) will be described with reference to the flowchart of FIG. 11. The data transmission control unit 32 of the slave station 30 monitors when the master station 20 transmits a polling message on the bus 50, and immediately receives the polling message when it is transmitted on the bus 50 ( SC1).

Next, the terminal address ADR is extracted from the received polling message in the format shown in FIG. It is compared with the address and determines whether the two match (SC3).

[0053]Then, both of them match, and currently,
If there is data to be sent to the main station 20, a response message in the format shown in FIG. 7 is immediately sent (replyed) to the main station 20 via the line driver 36 and bus 50 (SC6).

[0054] Then, an acknowledgment (ACK) is received within a predetermined time.
If received (SC7→SC8), the sending of the response message (the transmission data) to the polling from the main station 20 is completed (normal end).

[0055] Furthermore, if an acknowledgment (ACK) is not received within a predetermined time after transmitting the response message to the main station 20 (SC8), the response message to the main station 20 is not correctly transmitted. It is assumed that there was no
In order to resend the response message to the main station 20 again, the process returns to the process SC1 and waits for the next polling message to be sent from the main station 20.

[0056] Also, in the flowchart of the same figure,
Although not explicitly stated, if there is no data to be transmitted when the own station is polled, a message to that effect (for example, a message consisting of one predetermined character) is sent to the main station 20, or Or it does not respond to the main station 20 (no response).

On the other hand, in the above judgment process SC3, if the terminal address of the received polling memory is not equal to the terminal address of the own station, that is, it is not a polling message addressed to the own station, a delay time t1 is generated by random number generation, and the delay time t1 is generated by random number generation. During time t1, data on the bus 50 is read through the line receiver 36, and other slave stations 30
Monitoring is performed to see if any data is being sent to (SC4). If no data is received via the line receiver 36 during the time t1 (SC5 (N)), a response message is transmitted to the main station 20 (SC6).

[0058] Then, wait for the acknowledgment (ACK) to be received within a predetermined time, and if the acknowledgment (ACK) is received within the predetermined time, it is assumed that the data transmission to the main station 20 has been successfully completed. Determine and end the process (SC7→S
When C8(N)) fails to receive an acknowledgment (ACK) within a predetermined time, the response message is sent to another slave station 30.
It is determined that there has been a collision with the transmitted data, and the system waits for the reception of the polling message again (SC7 → SC8 (Y)
→SC1).

On the other hand, when any data is received during the above time t1, it is determined that there is a conflict with another slave station 30, and the state returns to waiting for reception of a polling message (SC5 (Y) → SC1). .

In the above operation, the length of the randomized time t1 is sufficiently longer than the time required for the slave station 30 to transmit one character via the bus 50, and the master station 20 performs polling. The maximum time for waiting for the reception of a response message from t0 (the time approximately equal to the polling cycle when there is no response message) is set to an arbitrary time length shorter than t0. That is, a random number that satisfies such conditions is generated and the above-mentioned time t1 is set.

In this embodiment, as described above, when the slave station 30 specified in the polling message (the slave station 30 polled by the master station 20) has data to be transmitted to the master station 20, By immediately transmitting a response message to the master station 20 when polled, data can be transmitted to the master station 20 with priority over other slave stations 30. Also, to the slave station 30 that was polled,
If there is no data to be transmitted to the main station 20, even the slave station 30 that is not polled will send the data to the main station 20 after an appropriate delay (equal to the random numbered time t1) from the polling time. It is possible to send data to.

In other words, in this embodiment, the polled slave station 30 can send data preferentially to the master station 20; If there is no data to be polled, the slave stations 30 that have not been polled can transmit data to the master station 20. In addition, the main station 20
The slave station 30 that has sent data (transmitted a response message) to the polling list is always connected to the end of the polling list.

Therefore, even if a large amount of data is exchanged between the master station 20 and slave stations 30, all slave stations 30
When the polling of other slave stations 30 is completed, it is possible to obtain the right to transmit data at approximately constant intervals. In addition, when there is little data exchanged between the master station 20 and the slave station 30, when the polled slave station 30 has no data to send to the master station 20, other unpolled slave stations 30 can send data to the master station 20, which allows the slave stations to send data to the master station much more quickly than in the conventional system where only polled slave stations can send data to the master station. (The throughput of data transmission from the slave station to the master station is improved). In addition, polling of the slave stations from the master station can be performed efficiently (the slave stations that are not transmitting data
0 is polled preferentially).

[0064]

[Effects of the Invention] According to the present invention, polling of slave stations is performed one by one in order, and if the polled slave station has no data to transmit to the master station, the data to be transmitted to the master station is Since a slave station that has not been polled can send data to the master station using the free time until the next polling, the amount of data exchanged between the master station and slave stations is small and the number of slave stations is small. Not only when the amount of data is large, but also when the amount of data exchanged between the master station and the slave station is large, each slave station can transmit data to the master station more quickly and efficiently than before.

[Brief explanation of drawings]

FIG. 1 is a functional block diagram of the present invention.

FIG. 2 is a system configuration diagram of a computer network.

FIG. 3 is a diagram showing the internal configuration of the main station.

FIG. 4 is a diagram showing the internal configuration of a slave station.

FIG. 5 is a flowchart illustrating the operation of the data reception control section of the main station.

[Figure 6] Format of polling message
FIG.

FIG. 7 is a diagram showing the format of a response message.

FIG. 8 is a flowchart illustrating polling list update processing.

FIG. 9 is a diagram illustrating a specific example of polling list update processing (Part 1).

FIG. 10 is a diagram illustrating a specific example of polling list update processing (Part 2).

FIG. 11 is a flowchart illustrating the operation of the data transmission control section of the slave station.

[Explanation of symbols]

1 Master station 2 (2-1, 2-2,...2-N) Slave station 2a
Monitoring means 2b Time generating means 3 Common transmission path

Claims (1)

[Claims] Claim 1: In a system in which one master station and a plurality of slave stations that transmit data to the master station are connected to one common transmission path, the master station polls each slave station one by one. In the transmission control method using polling, in which each slave station is invited to transmit data, each slave station has a time generating means that arbitrarily generates a time within a predetermined range, and a time generator that sends data onto the common transmission path. If there is data to be transmitted to the master station, each polled slave station immediately sends the transmitted data to the master station via the common transmission path. Each slave station that has not been polled monitors the data on the common transmission path by the monitoring means during an arbitrary time period generated by the time generation means when there is data to be transmitted to the master station. By polling, the transmission data is transmitted to the main station via the common transmission path when it is determined that no data has been sent from another slave station to the common transmission path for an arbitrary period of time. Transmission control method.