JPH0837541A - Data transmission system - Google Patents

Abstract

PURPOSE:To provide a data transmission system with which error detection can be more securely performed corresponding to the generation of error on a communication path and further proper data can be more securely transmitted to respective slave nodes by transmitting an error detected frame again. CONSTITUTION:This data transmission system is provided with an inspecting means 22 for inspecting the presence/absence of data error by providing a data sequence and an error check bit sequence for detecting data error added to the data sequence at the frame and by using the error check bit sequence for each received frame at the slave node connected to a ring-shaped network, error check bit sequence generating means 27 for updating the error check bit sequence for each frame and additionally outputting it to the next node, first scrambling means for loading first scramble to the data sequence and the error check bit sequence in the transmitting frame, and second scrambling means for loading second scramble different from the first scramble.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention, in a ring type network composed of one master node and a plurality of slave nodes, detects data error occurring on the network and retransmits the frame in which the error occurred. Regarding transmission method.

[0002]

[Prior art]

Reference name: IEEE Standard 802.5 Token Ring Access Meth Transistor Technology Special No. 8 "All about Data Communication Technology" p8-11 Parity check method for detecting communication errors in data transmission,
Constant mark code system, cyclic code (CRC) system and the like are well known.

Among them, the cyclic code (CRC) method is high in speed because it has a high check function for redundancy, is relatively resistant to burst noise, and has a simple CRC creation and detection circuit. It is often used for reliable data communication.

In the CRC method, a data string to be sent is divided by a generator polynomial, and the remainder (BCC: Block Check Charac
ter) is added as an error check bit string after the data string and transmitted, and the receiving side performs division using the same generator polynomial. If there is no remainder, the received data is judged to be correct. On the contrary, if there is a remainder, it is determined that the data string has an error.

Even in the token ring, which is a representative of the conventional ring-type network communication systems, an error during communication is detected by a CRC check. In the token ring, when an error during communication is detected by a CRC check, a bit called Ebit in the Ending Delimiter section in the frame is set to "1" and the system manager is notified of the error occurrence. The system manager recognizes the error occurrence by Ebit = "1" and performs necessary processing. Figure 8
It is a schematic diagram showing an example of a frame structure of a token ring. The bit indicated by E in the figure is Eb of the Ending Delimiter part.
It is a bit called it, and set this bit to "1",
Notify the system manager that an error has occurred. The system manager recognizes the error occurrence by Ebit = "1" and performs necessary processing.

Generally, in data transmission, a self-timing method for extracting timing from a received pulse sequence is widely used. In this case, in order to reduce the probability of occurrence of consecutive zero codes, a communication data sequence and a random sequence are mutually exclusive. A process called scramble (SCR) for converting data into different expressions by taking a logical OR is usually performed.

[0007]

In the token ring described above, the bit indicating an error is only 1 bit in the frame, and the Ending Delimiter part including Ebit is CR.
Since it is out of the C check target, a CRC error is detected at a certain node (called Station in token ring) and is transmitted with Ebit = "1", and then an error occurs and this bit changes from "1" to "0". There is a problem that the system manager cannot know the occurrence of the error when it is reversed to "".

Therefore, in the present invention, the scramble applied to the BCC bit string which is the error check bit string is CRC.
An object of the present invention is to provide a communication method in which an error due to a CRC check is surely propagated to a master node by switching to a different method depending on whether a check is normal or an error is detected.

[0009]

In order to solve the above problems, the present invention performs a data error check on a received frame at each node, and if normal, uses a regular scrambling method for a transmission data string and BCC. First scrambling means for scrambling, second scrambling means for scrambling the BCC different from the regular scrambling method when an error is detected, the transmission data string of the regular scrambled received frame and the BCC And descrambler means for returning the original. Here, the CRC check error is generated in all nodes after the node that detected the CRC check error, and the error occurrence is reliably transmitted to the master node. In addition, the master node generates an error when it detects the error. A means for recognizing the generated frame and a means for retransmitting the frame in which an error has occurred are provided.

[0010]

According to the present invention, when the received frame is normal as a result of the error check by the above means, the transmission frame to which a new error check bit string is added is normally scrambled by the first scramble means, and the next node In the above, the descrambler means restores the data as it was before the scramble, so the result of the error check in this node is normal unless an error occurs on the cable between the node and the immediately preceding node. On the other hand, if an error is detected in the received frame as a result of the error check, since the transmission frame to which a new error check bit string is added is scrambled differently from the regular scramble by the second scramble means,
In the next node, the descrambler means restores data different from that before scrambling, and as a result, an error is detected by an error check in this node, and an error is similarly detected in the following nodes,
The error can be reliably transmitted to the master node.
Further, the means for retransmitting the frame in which the error is detected on the network can surely transmit the correct data to each slave node.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 3 is a schematic configuration diagram of a ring network, FIG. 4 is an explanatory diagram of transmission / reception timing of frames in the case where there is no error, and FIG. 7 is a case where an error is detected in the network. 5 is an explanatory diagram of frame transmission / reception timing, FIG. 5 is a block diagram of the scramble circuit in FIG. 1, and FIG. 6 is an operation explanatory diagram of FIG.

FIG. 1 is a block diagram of a node on a ring network which realizes the communication system of the present invention, and FIG.
Are slave nodes 302 to 30 which are connected to the network and which receive and transmit data from the master node.
4 is a block diagram of FIG. 1 is an input unit from the network, 2 is an output unit to the network, and 3 and 4 are reception data output to a parallel transmission line (not shown) connected to the node and transmission data input from the parallel transmission line, respectively. is there. Reference numeral 20 denotes a frame synchronization unit that detects a synchronization bit string in a received frame before data transmission / reception in order to establish synchronization on the network. The operation inside the node is the synchronization detected by the frame synchronization unit 20. It is executed based on the timing. 21 is a descrambler for descrambling the received frame, 22 is a CR for detecting an error in the received frame
C check unit, in which a frame number information in a frame, a data unit, and BCC are divided by using a predetermined generator polynomial, and 12 is a CR
It is a CRC check result signal indicating the result of the C check unit 22, and outputs "1" when there is no error and outputs "0" when an error is detected.

Reference numeral 23 is a reception buffer for storing the received data and creating a predetermined delay, 24 is a transmission buffer for storing the data to be transmitted from the node itself, 29 is a data switching unit for inserting the transmission data from the node itself into the frame. In each of the figures, the data switching unit 29 receives the synchronization bit string, the command part, the frame number information, and the reception data which are input from the driver 21 to the reception buffer 23 and output from the reception buffer 23 according to the output timing of the transmission frame. The data output from the transmission buffer 24 and transmitted by the slave node itself is switched and output. Further, it is only necessary to include one timing generation circuit (not shown), whereby the reception timing is generated from the same reference signal as the transmission timing.

Reference numeral 27 denotes a CRC bit string generation unit, in which the frame number information and the transmission data string output from the data switching unit 29 are subjected to CRC calculation. That is, the frame number information and the transmission data string are divided by using a predetermined generator polynomial, which is the same as the division in the CRC check unit 22, and the remainder BCC is the transmission data as an error check bit string. Appended to columns. The CRC calculation may be performed on the synchronization bit string 200 and the command unit 201. 28 is a scramble circuit for scrambling the frame number information in the transmission frame and the transmission data string, but the command section 201 and the error check bit string 204 may also be scrambled. It should be noted that the control unit that generates various control timings and control signals based on the synchronization timing obtained by the frame synchronization unit 20 is not particularly shown in this configuration diagram.

FIG. 1A is a block diagram of a master node 301 that generates a frame that circulates on a network. According to the CRC check unit 22, the frame number generation unit 30 sequentially increments the frame number information if there is no CRC error in the received frame,
And generate a command. Also, the frame number generator 3
0 stores the frame number information of the latest frame which has circulated on the network and is normally received without a CRC error. When the CRC error is detected by the CRC check unit 22, based on the stored information, When a CRC error is detected, frame number information to be retransmitted and a command are generated. The data switching unit 25 generates the synchronous bit string 200 according to the output timing of the transmission frame from the transmission buffer 24, and thereafter reads from the command unit 201, the frame number information 202, and the reception buffer 23 from the frame number generation unit 30. The reception data of another node and the transmission data of the master node itself from the transmission buffer 24 are switched and output. Further, a timing generation circuit that generates a reception timing and a timing generation circuit that generates a transmission timing are separately provided, and the reception timing and the transmission timing are generated independently of each other. It should be noted that the master node and the slave node do not have to be of different types as devices, and the node can be arbitrarily set as either the slave or the master to switch the operation of some functional blocks.

Next, FIG. 2 shows an embodiment of a frame format for circulating a communication path according to the present invention. A method for establishing synchronization on a network will be described with reference to FIG. In order to detect the synchronization bit string by the frame synchronization unit 20 in each slave node, the master node transmits a synchronization establishment frame for frame synchronization based on its own timing. This synchronization establishment frame is shown in FIG.
As shown in (a), it includes a synchronization bit string 200 and a command section 201 containing a synchronization establishment command indicating that it is a synchronization establishment frame. However, as shown in FIG. 2B, frame number information 202 and a data section are included. It may further include 203, in which case the data area may be present and the data may be empty. The error check bit string 204 in the synchronization establishment frame is for checking whether the command unit 201 is correct, and may not be provided.

Receiving the synchronization establishment frame, each slave node can detect the start point of the frame by detecting the synchronization bit string 200, establish the frame synchronization, and generate various control timings. Further, it is possible to know that the received frame is the synchronization establishment frame by the synchronization establishment command of the command unit 201 following the establishment of the synchronization. This synchronization establishment command is used to indicate to each slave node that it is in a communication standby state, for example, a communication frame comes after this synchronization establishment frame. On the master node,
The synchronization establishment frame returned after circulating through the network enables itself to generate control timing on the receiving side. Then, the master node starts data communication and transmits a communication frame for data transmission / reception in each slave node.

Further, as shown in FIG. 2B, the communication frame has a sync bit string 20 for frame synchronization.
0, a command section 201 containing a communication mode command indicating a communication frame, frame number information 202,
An error check consisting of a data section 203 in which data areas (203-1 to 20-4) for four nodes are allocated as fixed channels, frame number information 202, and a BCC generated as a result of CRC calculation performed on the data section 203. Frame number information 2
02 occurs only on the master node, and usually if no data error is detected during network patrol,
The frame number incremented by 1 is transmitted from the master node. In the present embodiment, although the data areas for the four nodes (203-1 to 4-3) are allocated as fixed channels in the data section 203, the frame length does not necessarily have to be a fixed length. It may differ depending on the content.

FIG. 5 is a block diagram of the scramble circuit 28, and FIG. 6 is a schematic diagram for explaining the relationship between the operation timing and the frame. Scramble circuit 28
Includes a first scramble circuit 501 that receives the output 11 of the CRC bit string generation unit 27 and performs scrambling, and a second scramble circuit 502 that receives the output 11 and performs scrambling by a method different from that of the first scramble circuit 501. , CRC check result signal 12 is input, whereby a selector 504 for switching and selecting the output 511 from the first scramble circuit 501 and the output 512 from the second scramble circuit 502, and control from the frame synchronization unit 20. The selector signal 13 generated from the timing control section operated by
As a result, the output 515 to which the synchronous bit string 200 in the frame is sent and the output 514 from the selector 504.
And a selector 505 for switching and selecting.

In this example, the CRC calculation is performed by the synchronization bit string 200, the command part 201, and the frame number information 202.
And the data section 203 are scrambled, and the scrambling is performed by the command section 201, the frame number information 202, and the data section 2.
03 and the error check bit string 204, the operation of the scramble circuit 28 will be described. Of the outputs of the CRC bit string generator 27, the synchronous bit string 200 is not scrambled and is output from the selector 505 as the output 515 when the selector signal 13 is “1”. The scramble circuits 501 and 502 are reset while the synchronous bit string 200 is input, and the command section 201, the frame number information 202, and the data section 20.
It operates while 3 and BCC204 are input. Command part 201, frame number information 202, data part 2
03 and BCC 204 are scrambled, and when the selector signal 13 is “0”, the network output unit 2
Is output on the network.

At this time, if there is no data error in the CRC check, the CRC check result signal 12 becomes "1", and the output 511 is selected by the selector 504, as shown in FIG.
As shown in (a), the output 2 is followed by the sync bit string 200, and the command section 201, the frame number information 202, the data section 203, and the BCC 204 scrambled by the scramble circuit 501 are output. Conversely, CR
When the C check error is detected, the CRC check result signal 12 becomes “0”, and as shown in FIG. 6B, following the synchronization bit string 200, the command section 201 scrambled by the scramble circuit 501, the frame number information 202, the data After the section 203, the scramble circuit 502
The BCC 204 scrambled by is output.

4 and 7 show the transmission / reception timing of frames on the network. FIG. 4 shows CR.
When there is no C error, FIG. 7 shows the state of transmission and reception when an error is detected.

The communication procedure when there is no data error will be described with reference to FIG. In FIG. 4, first, the first frame from the master node (frame number information 202 = 0)
Will be sent. At this time, in the data portion 203 on the frame output from the master node, only the data transmitted by the master node exists. In the slave node 302 receiving this frame, after undergoing the above-mentioned internal operation,
The transmission data from the slave node 302 itself is inserted into the own node data area, and the frame is sent to the next slave node. Hereinafter, the frame in which each transmission data is inserted in the data area of each node in the same procedure in each node circulates to the master node 301. Since the slave node scrambles the error check bit string after judging the CRC check result, a delay of a certain time corresponding to the error check is provided until the data string input to the reception buffer 23 is read. .
In addition, in this embodiment, the master node 301 receives the frame from the slave node 303, and waits for the CRC check before the next frame (frame number information 20
2 = 1) is transmitted (FIGS. 4A and 4B).

Next, the communication procedure when a data error occurs will be described with reference to FIG. Now, in the frame whose frame information number is “N”, the node 302 and the node 30
If an error occurs on the communication path between nodes 3,
In 03, the CRC check error of the received frame is first detected (FIG. 7 (a)). In the node 303 that has detected the CRC check error, as shown in FIG.
The operation shown in (b) is executed, and the BCC scrambled by the scramble circuit 502 is sent to the error check bit string of the transmission frame (FIG. 7 (b)).

The node 304 that has received this frame descrambles the received frame. As described above, the descramble circuit 21 only scrambles the data scrambled by the scrambler 501. For the scrambled data other than the original data, random data different from the original data is output. Therefore, the descramble execution by the node 304 restores the frame number information and the original data for the data part, but random data is output to the error check bit string, and as a result, the CRC check part is output. A data error is detected by 22. Similarly, the error check bit string of the frame output from the node 304 is scrambled by the scramble circuit 502 (FIG. 7C), and finally the master node 301.
The data error will be propagated up to.

Master node 3 which has detected a data error
01 recognizes the frame number N of the received frame so that the frame in which the data error has occurred is the frame number N, and the data that should be transmitted / received in this frame is at least one of the nodes on the ring network. Know that it was not received correctly. At this point, the master node has already transmitted the frame with the frame information number = N + 1, so that the frame number of the frame to be transmitted next is N and the retransmission is executed (FIG. 7 (d)).

As for the frame with the frame information number = N + 1, the error is detected in the frame with the frame number = N before the transmission of the error check bit string is started. Scramble circuit 5 for the check bit string
02 can be scrambled. For example, when continuity of the order of transmission data is required like image data, the error is propagated to the frame with the frame information number = N + 1 so that each node can receive the frame N before receiving the frame N. It suffices not to receive N + 1.

The slave node can retransmit the data transmitted in the frame in error by transmitting the data corresponding to the received frame number.

[0029]

As described above, according to the present invention, more reliable error detection can be performed with respect to the occurrence of an error in a communication path, and further, by retransmitting a frame in which an error has been detected, more reliable error detection can be performed. It is possible to realize a highly reliable data transmission method capable of transmitting correct data to each slave node.

[Brief description of drawings]

FIG. 1A is a configuration diagram of a master node on a ring network that realizes a communication system of the present invention. (B) It is a block diagram of the slave node on the ring network which implement | achieves the communication system of this invention.

FIG. 2 is a schematic diagram showing an embodiment of a frame format according to the present invention that circulates over a communication path.

FIG. 3 is a schematic configuration diagram of a ring network in the present invention.

FIG. 4 is a conceptual diagram showing a frame transmission / reception timing on a ring network when there is no error according to the present invention.

FIG. 5 is a block diagram of a scramble circuit unit according to the present invention.

FIG. 6 is a conceptual diagram showing operation timing of a scramble circuit diagram according to the present invention.

FIG. 7 is a conceptual diagram showing a frame transmission / reception timing on a ring network when an error is detected according to the present invention.

FIG. 8 is a schematic diagram of a general frame of Token Ring.

[Description of Codes] 1 Input Unit 2 Output Unit 3 Received Data Output 4 Transmitted Data Input 12 CRC Check Result Signal 20 Frame Synchronous Unit 21 Descramble Circuit 22 CRC Check Unit 23 Receive Buffer 24 Transmit Buffer 25, 29 Data Switch Unit 26 Delay Circuit 27 CRC bit string generator 28 Scrambling circuit 30 Frame number generator 101 Node 200 connected to network 200 Synchronous bit string 201 Command part 202 Frame number information 203 Data part 204 Error check bit string 501,502 Scrambling circuit 504,505 Selector 13 Selector 505 Selector signal 301 Master node 302 to 304 Slave node

Claims (5)

[Claims] 1. A ring network is formed by one master node and a plurality of slave nodes, and a frame circulating in the ring network detects a data sequence and a data error added to the data sequence. The slave node, which has an error check bit string and is connected to the ring network, inspects a data error for each received frame using the error check bit string, and an error check for each frame. A data transmission method comprising: an error check bit string generating means for updating and adding a bit string to output to a next node; and a first scramble means for first scrambling a data string and an error check bit string in a transmission frame, The slave node is the first A descrambling means for restoring the data sequence and error check bit sequence of the received frame subjected to the scrambling, and a second scramble for applying a second scramble different from the first scramble to the data sequence and the error check bit sequence in the transmission frame. A scrambler, and when the data of the received frame is determined to be normal by the checker, the first scrambler scrambles the data string and the error check bit string in the transmit frame, A data transmission method characterized in that, when a data error is detected from the received frame by the inspection means, the second scramble means applies the second scramble to the data sequence and the error check bit sequence in the transmission frame. 2. The data transmission method according to claim 1, wherein the error check bit string is generated by a cyclic code method. 3. The method according to claim 1, wherein the frame in which the data error is detected is sequentially propagated to another node to notify the master node of the occurrence of the data error, and the frame is retransmitted from the master node. Data transmission method. 4. The data transmission method according to claim 1, wherein the frame that circulates in the ring network is
Each node includes frame number information indicating the order of frames transmitted, and each node detects / recognizes the frame number of the received frame, and the first, second scramble, and / or descramble are added to this frame number. A data transmission method characterized by being applied. 5. The data transmission method according to claim 1, wherein the master node, when receiving a frame in which a data error is detected, sets the frame number of the frame as a frame number of a frame to be transmitted next. Data transmission method characterized by.