JPH0683236B2 - Transmission line control method for loop network - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial field of application) The present invention promptly operates in a loop network when an error occurs in transmission / reception control on a transmission line due to a code error on the transmission line. The present invention relates to a transmission line control method capable of performing normal transmission / reception control.

(Prior Art) As a communication method for realizing the integration of broadband services such as video communication and high-definition images in addition to the integration of voice communication / data communication, for example, 1983 IEICE Exchange Study Group Material SE83-107. There is a loop network described in "Study on high-speed synchronous packet loop" described above.

This loop network constitutes the network as shown in FIG. In the figure, the trigger station TS20 sends a polling signal onto the transmission line 24 to perform polling control on the node stations NS21, NS22, NS23. Also, node stations NS21, NS22, NS23
Performs packet transmission / reception via the transmission line 24 in accordance with the polling control. Each of these stations is a transmission line
By 24, they are connected in a loop as shown. (In the present specification, the trigger station is
S, the node station is abbreviated as NS, and TS and N
S is generically called a station. ) In this loop-shaped network, as shown in FIG. 7, TS20 sends a polling signal (p1 or p2), and each NS monitors the polling signal passing through the transmission path 24 and sends packets. After preparing and acquiring the transmission right, the packet is transmitted. For example, in the example of FIG. 7, the NS21 receives the polling signal p1 transmitted by the TS20, acquires the transmission right after passing the signal, and transmits the packet a1. NS22 receives packet a1 after receiving polling signal p1.
After that, the transmission right is acquired and the packet b1 is transmitted. Similarly, NS23 also transmits packet c1.

When the polling signal p1 transmitted by the TS20 makes one round in the transmission line 24 and returns, the TS20 transmits the polling signal p1 to the transmission line 24.
After removing from 24 (signals removed at each station in the figure are indicated by *), the packets a1, b1 and c1 transmitted by NS21, NS22 and NS23 are passed, and then the transmission right is acquired. The polling signal p2 is transmitted, and the transmission of the packet a2 is started to NS21. Also, each NS also removes from the transmission line 24 when the packet transmitted by itself returns around the transmission line 24. In this way, the transmitted signal is removed at the transmission source station after passing through the transmission path 24 once.

(Problems of Prior Art) Each station in the loop network is
When the transmission of a signal is started, the transmission line 24 is temporarily disconnected, the signal in its own station is transmitted, the signal waits for one round in the transmission line 24, the signal is removed from the transmission line 24, and then transmitted again. Connect line 24. In this way, each station controls the opening / closing of the transmission path 24 during transmission. Therefore, the transmission line 24
If a station mistakenly opens or closes the transmission line 24 due to a code error in the above, the transmitted packet may be removed at the station before it reaches the receiving station, or the packet may be transmitted. Despite going through 24 cycles, a failure occurs such as not being removed at the transmission source station, causing an abnormality in the access method of each station. For example, in the example of FIG. 8, when the end delimiter indicating the end of the packet appears due to a code error in the middle of the packet a1 transmitted by the NS21, the NS21 that transmitted the packet a1 is When the packet has returned through the transmission path 24 and returned, the removal of the packet a1 ends when the end delimiter caused by the code error is detected. Therefore, as indicated by the diagonal lines in FIG. 8, a part of the latter half of the packet a1 is not removed and remains.

On the other hand, NS22 sends the remaining signal to the packet
The transmission line 24 is cut and removed until it is identified as b1 and the end delimiter of the signal indicated by the diagonal lines is detected, and the transmission line 24 is returned to the passing state when the end delimiter is detected. Therefore, even if the packet b1 transmitted by the NS22 goes around the transmission path 24 and returns to the transmission source station NS22, it remains without being removed at the same station and is removed at NS23. In the same way
The packet c1 sent by NS23 is removed at TS20, and TS
The polling signal p2 transmitted by 20 is removed in NS21, and thereafter, an abnormal state in which each source station removes a signal different from the packet transmitted by itself, continues.

(Object of the Invention) The object of the present invention is caused by a code error on a transmission line,
It is an object of the present invention to provide a transmission path control method capable of promptly performing normal transmission / reception control even when a signal that is not removed even after one round of the transmission path is generated and an abnormality occurs in transmission / reception control of each station.

(Structure of the Invention) According to the present invention, it is composed of one trigger station and a plurality of node stations, each station is connected in a loop by a transmission line, the trigger station transmits a polling signal to the transmission line, A transmission path control method of a loop network for controlling transmission to the node station, wherein the trigger station comprises a switch provided in series with the transmission path to open and close the transmission path, and a loop of the transmission path. And a timer for notifying the passage of a fixed time determined based on a propagation time, the trigger station disconnects the transmission path at the time of transmitting the polling signal to put it in a transmission state, and activates the timer to set the fixed time. The transmission line is kept disconnected until the timer is recognized, According to the transmission path control method of the loop network, the transmission path is set to the passing state when the end of the polling signal that has passed through the transmission path is detected after the passage of the fixed time.

(Detailed Description of Configuration) A transmission path control system of a loop network according to the present invention will be described in detail with reference to the drawings. The polling signal transmitted by the TS20 constituting the loop network and the frame configuration of the packet transmitted by each NS are secondly described.
Shown in the figure. These signals are the header part (HDR) that contains the address of the source station of the signal, the address of the receiving station and the control information of the frame, and polling information for each NS, and the message part (INF
O), and before and after this, a start delimiter (SD) and an end delimiter (ED) indicating the start and end of the signal are added,
Packetized as shown in FIG. 2 and transmitted on the transmission line 24. In this way, when the TS20 and each station of each NS end the transmission, the end delimiter (ED) is sent to end the transmission.

Further, the signal first received from the upstream side of the transmission line 24 after each station finishes the transmission is the signal transmitted by itself after one round of the transmission line 24 in a normal state. The TS20 receives the polling signal p1 that it sent first after the end of transmission, and then NS21, NS22,
Each of the NSs 23 receives the packets a1, b1, c1 that it has transmitted first after the end of transmission. Therefore, each station can recognize that the end delimiter (ED) detected first by monitoring the transmission line 24 after the end of transmission is the signal indicating the end of the packet transmitted by itself, and the removal of the packet transmitted by itself is It is performed until the end delimiter (ED) is detected after the end of transmission.

Furthermore, the TS20 starts polling control to remove these signals in order to deal with the case where there are signals that are not removed even after one round of the transmission line 24 due to the above-mentioned code error on the transmission line 24. After that, the transmission line 24 is kept in the disconnected state for a certain period of time, and even if the end delimiter (ED) is detected within the certain period of time, the transmission line 24 is not returned to the passing state.

FIG. 1 is a diagram showing a configuration of the TS20 according to the present invention,
In the figure, the signal on the transmission line 24 is bypassed to the adjacent NS21 via the switch 1 when the TS20 is not transmitting, and the ED detection unit 2 for detecting the end delimiter (ED) of the packet, and It is input to the transmission right acquisition unit 3 which monitors the state on the transmission path 24 and controls acquisition of the transmission right. The polling control unit 5 stores the polling signal in the POL register 7 by the output 5-2 of the same unit, and instructs the access control unit 4 to start the polling control by the output 5-1 of the same unit. When transmitting the polling signal on the transmission line 24, the access control unit 4 connects the switch 1 to the output terminal of the POL register 7 in which the polling signal is stored by the output 4-1 of the same unit, and outputs the polling signal. Control transmission. At the same time, the access control unit 4 initializes and activates the tire 6. The timer 6 measures the elapse of a fixed time specified based on the round-trip propagation time of the transmission path 24.

(Embodiment) An embodiment of the present invention will be described using the configuration of the TS 20 shown in FIG. 1 and the signal flow on the transmission line 24 shown in FIGS. 3 and 4.

The polling control unit 5 in the TS20 stores the polling signal in the POL register 7 by the output 5-2 of the same unit, and instructs the access control unit 4 to start the polling control by the output 5-1 of the same unit. When receiving the polling control start instruction, the access control unit 4 sets the output 4-1 to the active (active) state, connects the switch 1 to the output terminal of the POL register 7, and starts transmitting the polling signal. After the polling control unit 5 of the TS20 starts polling control as described above, the access control unit 4 transmits a polling signal every time the transmission right acquisition unit 3 acquires the transmission right.

The access control unit 4 initializes and activates the timer 6 at the same time as the transmission of the polling signal is started. As shown in FIG. 3, the timer 6 measures the elapse of a fixed time t1 determined based on the round trip time of the transmission path 24. The value of this t1 is a predicted value of the time that the polling signal transmitted by the TS20 makes one round in the transmission line 24, and is equal to the sum of the round trip propagation time of the transmission line 24 and the polling signal transmission time, or slightly smaller than that. Shall take a small value. (Details of the value of t1 will be described later.) The output 4-1 of the access control unit 4 remains active (active) until the timer 6 notifies the elapse of the predetermined time t1 after the transmission of the polling signal is started. , So switch 1
Keeps the transmission line 24 disconnected and removes the signal on the transmission line 24. When the end delimiter (ED) is detected by the ED detection unit 2 after the elapse of the fixed time t1 from the timer 6, the output of the AND gate 8 becomes active (active) and the access control is performed. Upon receiving this, the unit 4 puts the output 4-1 of the unit 4 into the inactive state.

When the output 4-1 of the access control unit 4 becomes inactive (inactive), the switch 1 is connected to the transmission path 24 side, and subsequent signals are bypassed to the adjacent NS21.

Therefore, as described above with reference to FIG. 8, due to a code error on the transmission line 24, an end delimiter (ED) occurs in the middle of the packet, and the transmission / reception control thereafter becomes abnormal. Even if a signal that is not removed is generated even if the TS20 does not return the switch 1 from the disconnection state to the passing state even if the end delimiter (ED) is detected until the constant time t1, the constant time t1 Does not return the switch 1 to the passing state based on the detection of the end delimiter (ED) of the signal that has passed before the passage of, ie, the signal that has not been removed by the NS, and is received after the elapse of a certain time t1. When the end delimiter (ED) of the polling signal is detected, the switch 1 is returned to the passing state. Therefore, in TS20, in addition to the polling signal transmitted by itself, the remaining signals not removed by each NS are removed from the transmission path 24, and the normal state is restored.

FIG. 3 shows an abnormal state as described in FIG. 8, that is, an end delimiter (ED) occurs due to a code error in the transmission line 24 in the middle of the packet a1 transmitted by NS21, and the subsequent removal control becomes abnormal. Then, when TS20 receives the packet c1 that was not removed by NS23 after the end of transmission of polling signal p2 and detects the end delimiter (ED) of the packet, TS20
20 shows a signal flow on the transmission line 24 when the transmission line control according to the present invention is executed.

The TS20 does not return the transmission line 24 to the passing state during the constant time t1 shown in FIG. 3 and keeps the transmission line 24 in the disconnected state, and the NS23 left on the transmission line 24 during the constant time t1. The transmitted packet c1 is removed. Then, the TS 20 returns the transmission line 24 to the passing state at the time when the end delimiter (ED) detected after the elapse of the certain time t1, that is, the end delimiter (ED) of the polling signal p2 transmitted by itself is detected. As described above, in the prior art, the polling signal is not removed in TS20 at the time of failure, whereas in the present invention, the polling signal is TS.
It is removed at 20, after which a new polling signal is transmitted, and thereafter, the normal transmission / reception state can be restored.

Note that the value of t1 is determined in consideration of the operation accuracy of the timer 6 itself, and when the polling signal transmitted by the TS20 makes a round in the transmission path 24 and returns after the elapse of the fixed time t1, the end delimiter ( ED) is detected and the transmission line 24 can be connected in a passing state. For example, let td be the round-trip propagation time of the transmission path 24, and tp be the time required to transmit the polling signal.
Then, the value of the measurement time t1 of the timer 6 is t1 = td
+ Tp However, considering the accuracy of timer 6, t1 <
It may be td + tp.

The set value of t1 will be described using a specific numerical example. When the transmission speed of the transmission line 24 of the loop network is 32 Mbps and the length of the polling signal is 240 bits, the time tp required for transmitting the polling signal is 240 bits / 32 Mbps = 7.
5 μsec. Now, it is assumed that the round trip delay time td of the transmission path 24 is 48.46875 μsec (the length of 1551 bits) including the delay in the station. In this case, t1 = 48.46875 + 7.5
It may be set to μsec, but the measurement accuracy of timer 6 is 1 of tp
It may be roughened to a unit of 4 μsec of / 2 and set as t1 = 48 + 4 = 52 μsec.

The value of this td is, for example, patent application number Sho 58-055438.
Specification "Communication control method in loop network"
Can be measured according to the method described in.

The access control unit 4 in the TS20 can also initialize and start the timer 6 at the end of transmission of the polling signal. FIG. 4 is a diagram showing the state of signals on the transmission line 24 at this time. . The fixed time t2 set in the timer 6 is t2
≤td, and the value is set in consideration of the measurement accuracy of the timer 6 as in the previous case.

Further, each station forming the loop network may be required to transmit a plurality of packets after acquiring the transmission right. In this case, the patent application number Sho
As described in No. 59-084581, "Communication control method in loop network", indicating that the last packet sent by each station is the same as the last packet sent by each station. A delimiter may be added, and opening / closing control of the transmission line 24 may be performed based on the detection of the end delimiter. FIG. 5 shows an example in which a station transmits three packets in succession. As shown in FIG. 5, this station confirms that the same packet is the last packet in the third packet. An end delimiter (LED) shown is added, and an end delimiter (IED) indicating that the packet is in the middle of transmission is added to the first and second packets for transmission. Therefore, based on the detection of the end delimiter (LED), the end of the signal transmitted by itself is identified and the removal control of the signal is performed.

(Effect of the Invention) As described above, the trigger station (TS) forming the loop network removes the signal on the transmission line by transmitting the polling signal and then disconnecting the transmission line for a certain period of time. Therefore, due to a code error that occurs during transmission, a signal that is not removed even if it makes a round in the transmission line occurs, and even when an abnormality occurs in transmission / reception control, the effect is limited to the polling control in which the code error occurs, The normal state can be recovered from the next polling control.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration in a trigger station TS20 that constitutes a loop network, FIG. 2 is a diagram showing a frame configuration of a signal transmitted by each station, and FIGS. 3 and 4 are a signal flow and FIG. 5 is a diagram showing the relationship between the round trip propagation time td of the transmission path 24 and the set values t1 and t2 of the timer 6, FIG. 5 is a diagram showing the frame structure of a signal when a plurality of packets are transmitted, and FIG. 7 and 8 are diagrams showing the structure of the network, and show the flow of signals on the transmission line 24. In the figure, 1 is a switch, 2 is an ED detection unit, 3 is a transmission right acquisition unit, 4 is an access control unit, 5 is a polling control unit, 6 is a timer,
7 is a POL register, 8 is an AND gate, and 24 is a transmission line. 20 is a trigger station (TS), 21 and 22 and 23
Indicates a node station (NS).

Claims (1)

[Claims] 1. A trigger station and a plurality of node stations, each station being connected in a loop by a transmission line, the trigger station transmitting a polling signal to the transmission line, and to the node station. A transmission path control method of a loop network for controlling transmission, wherein the trigger station is determined based on a switch provided in series with the transmission path for opening and closing the transmission path, and a round trip time of the transmission path. A timer for notifying the elapse of a certain period of time, the trigger station disconnects the transmission path at the time of transmitting the polling signal to put it in a transmission state, and activates the timer until it recognizes the elapse of the certain period of time. Hold the transmission line in a disconnected state, and use the timer for the fixed time. After the detection, upon termination detection of said polling signal cycle through the transmission path, the transmission path control method for a loop-like network, characterized by setting the transmission path to the passing state.