JPS5915427B2 - Error correction method for two-wire time-division bidirectional transmission - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an error correction method that enables error correction when an error occurs in digital communication based on a time-division direction controlled bidirectional transmission method.

One of the methods for transmitting digital bidirectional information over a two-wire line
One of them is the time-division direction control type bidirectional transmission system (for example, see IEICE Communication System Study Group material ``Study on the method of configuring an all-digital telephone subscriber system'' CS78-20).

Figure 1 shows the configuration of this conventional time-division directional control type bidirectional transmission system, in which the connection terminal 11 with the two-wire line is a balanced-to-unbalanced conversion transformer that connects the two-wire line and the four-wire line. 1
2 to a transmitter 13 for sending signals to the two-wire line and a receiver 14 for receiving signals from the two-wire line. The signal from the receiver 14 is stored in the reception temporary storage device 15.
The signal is temporarily stored in the 4-wire side output terminal 16 and sent to the 4-wire side output terminal 16. The information on the 4-line side receiver terminal 17 is temporarily stored in a transmission temporary storage device 18 and supplied to the transmitter 13.

For example, the input signal to the input terminal 17 is as shown in FIG. 2A, the transmission/reception signal at the two-wire line connection terminal 11 is as shown in FIG. 2B, and the output signal of the output terminal 16 is as shown in FIG. 2C, respectively.
Frame period Tf) Burst period Tb) Frame period T
Transmission signals 5i (i=0, 1, 2, 3) separated by f,
Received signal Ri (i=0,
1, 2, 3). Suppose now that nkb/s is transmitted and nkb/s is received at the same time. The transmission signal at the input terminal 17 (FIG. 2A) is divided into frame periods Tf, and S
It is divided into i blocks and written to the transmission temporary storage device 18. The written data is read out from the storage device 18 at a speed faster than 2n (speed=1/Ts) and sent to the transmitter 13. The receiving operation is performed in the reverse manner. In this way, bidirectional transmission using a two-wire line becomes possible. In this prior art, terminals 17 and 11
Since the input signal and transmission line transmission signal are each transmitted only once as shown in FIGS. 2A and 2B, it is impossible to perform error correction unless an error correction code is used. However, the error correction code decoder has the disadvantage that it is complicated to implement and has a large number of redundant bits for error correction. In order to eliminate these drawbacks, the present invention takes advantage of the fact that the time-division direction controlled bidirectional transmission system is originally half-duplex communication and requires a temporary storage device, and uses an error detection code. The configuration is such that retransmission is performed when an error is detected.

Error detection codes have fewer redundant bits than error correction codes and are much easier to detect. Therefore, this method allows error correction with simpler hardware than when error correction codes are used. FIG. 3 is a diagram showing the basic principle of the error correction method according to the present invention. Figure 3A shows the transmission signal at the two-wire line connection terminal 11 (Figure 1) when using the conventional method, where Tf is the frame period and Si
(1=1, 2, 3, 4) is a transmission signal divided by the frame period Tf. Since Si contains information from the input terminal 17, this is defined as an information burst signal. As shown in FIG. 3A, in the conventional system, N (four in the figure) information burst signals S exist during the NXTf time. FIG. 3B shows the transmitted signal at the two-wire line connection terminal 11 when using the error correction method according to the present invention, where Tf' is the frame period and Si (1=1, 2
, 3, 4) are information burst signals separated by the frame period TC, and Di (1=1) is a dummy burst signal that does not include information from the input terminal 17.

As shown in FIG. 3B, during (N+k)×Tf' time, N (four in the figure) information burst signals S and k (
In the figure, there is one dummy burst signal D. NX
Tf-(N><k)×Tf'...If (1) is satisfied, the transmission information of NKb/s will be transmitted in NKb/s on the receiving side.
It can be received at FIGS. 3C and 3D respectively show the transmitted signal and received signal at the two-wire line connection terminal 11 when an error actually occurs on the transmitting side when using the error correction method of the present invention, -1, 2, 3
, 4) is the transmission information burst signal, Ri (1=1, 2, 3)
, 4) are received information burst signals, and Di (1=1) is a dummy burst signal. Each transmitted information burst Sil and received information burst Ri includes an error detection bit that detects whether or not each information burst signal is erroneous, and a retransmission request bit that requests retransmission to the transmitting side when an error is detected. include. If an error is detected while receiving N information burst signals Si (1=1, 2, 3, 4) on the receiving side, in FIG. 3C, the first transmitted information burst signal S2
If an error is detected while receiving the next received information burst signal Ri (1=1, 2, 3, 4), the retransmission request bit in R2 in FIG. 3D is used to request a retransmission to the transmitting side. do.
Upon receiving this, the transmitting side retransmits the information burst signal in which the error was detected, S2 in this example, instead of transmitting the next information burst signal, S3 in FIG. 3C. Since there are k dummy burst signals for N information burst signals, retransmission can be performed up to k times (once in the figure) within (N+k) XTC time. Fig. 4 shows the format of the burst signal, in which the information bit 21, the error detection bit 22
, a retransmission request bit 23. In Figure 3, (N+k
) × Tf' time, the dummy burst signals are arranged at the end of the information burst signals, but due to the agreement between the transmitter and the receiver, they can be placed at any position within the (N+k)Tf' time, and the retransmission request bit 23 can be retransmitted. It is of course possible to use other methods of displaying the request, such as changing the transmission code (bipolar violation, etc.), changing the frame pattern, etc. FIG. 5 shows an embodiment of the error correction system according to the present invention, and to simplify the explanation, the case where N=4 and k=1 will be described, but it goes without saying that it can be applied to general N and k. For reasons explained later using FIGS. 7 and 8, when N=4 and k=1, three transmission temporary storage devices 18 and three reception temporary storage devices 15 are required. Storage part 18a in 18
, 18b, 18c are stored in the storage unit 1 in the reception temporary storage device 15.
5a, 15b, and 15c are provided, respectively. Storage part 1
8a, 18b, 18c are write address designation counters 2
4, the switch 25 is controlled to be connected to the input terminal 17. Further, a switch 27 is controlled by a read address designation counter 26 to
5c is switched and connected to the output terminal 16. The read address designation counter 28 is activated by the switch 31 through the inhibit circuit 29.
is controlled by the storage section 18 through the switch 31.
a, 18b, and 18c are switched and connected to the error detection circuit 32. The retransmission request detection circuit 33 detects a retransmission request signal from the output reception signal of the receiver 14, and the retransmission request detection circuit 33
The inhibit circuit 29 inhibits the output of the read address designation counter 28 based on the detected output. A retransmission request signal input circuit 34 switches and connects the output side of the retransmission request signal generation circuit 35 to the transmitter 13 from the output side of the error detection encoder 32, and inputs the retransmission request signal into the transmission burst signal. A retransmission request signal separation circuit 36 separates the retransmission request signal from the received burst signal output from the receiver 14 and supplies it to the retransmission request detection circuit 33 . The output of receiver 14 is provided to error detection decoder 37 and switch 38. Error detection decoder 3
The prohibition circuit 39 is controlled by the control of the prohibition circuit 39.
The output of the write address designation counter 41 controls the switch 38 through the memory section 15a, 15.
Switch and connect to b and 15c. The input signal to the input terminal 17 in FIG. 5 is, for example, the Al 2-wire line connection terminal 11 in FIG.
The transmitted and received signals in FIG. 6 are the output signals of the B1 output terminal 16.
The respective states of D, E, and Fl reception temporary storage devices 15a, 15・B, and l5c are shown in FIG. 6.
The result will be as shown in Figures G, H, and I.

Write status to each temporary storage device to Wrl read status Re
It is shown in From equation (2) described later, transmission of the transmission information burst S1 is started from the two-wire line connection terminal 11 2Tf' after inputting the transmission information S1 at the input terminal 17. Furthermore, since q = O from the equation (8) below, there is a decoding delay of MTf from the equation (9) below.

The transmission signals S, , S2, S3, and S4 shown in FIG. 2 after adding a retransmission request signal using the retransmission request input circuit 34 and the retransmission request signal generation circuit 35.
It is sent as a burst signal to the wire line. In this example, since no error occurred, there is no need to retransmit, and therefore the dummy burst signal D is used in the last frame period TC.
Sends 1. Received burst signals Rl, R2, R3, R
4 is sequentially input to a retransmission request signal detection circuit 33 and an error detection decoder 37 via a receiver 7 under the control of a retransmission request signal separation circuit 36.

When the retransmission request signal is detected by the detection circuit 33, the output of the counter 28 is inhibited, and reading from the same temporary transmission storage device is enabled again. The input to the error detection decoder 37 is written to one of the reception temporary storage devices 15a, 15b, 15c. For example, the received signal R1 is written to 15a. At this time, if the error detection and decoding circuit 37 does not detect an error, the retransmission request signal generation circuit 35 does not issue a retransmission request signal, and after a certain period of time, the signal R is sent from the reception temporary storage device 15a to the output terminal 16. Ru. The next received signal, the first signal R2, is written to the storage device 15b. If an error is detected at this time, the error detection decoding circuit 3
7 inhibits the output of the counter 41 and starts the next frame period T.
It is arranged so that it can be stored in the storage device 15b even when ν.
On the other hand, immediately after an error is detected, the retransmission request signal generation circuit 35
If a retransmission request signal is input into the transmission information burst S3 using
2 is retransmitted, and this second signal R2 is sent to the storage device 15b.
written to. Since k=1, retransmission is limited to one time. Therefore, the subsequent signals R3 and R4 are sequentially stored in the storage device 15c.
, 15a and sent to the output terminal 16.
FIG. 7 is a diagram showing the number of transmission temporary storage devices, transmission timing, reception timing, and decoding delay time for realizing the error correction method according to the present invention.

The input signal at the input terminal 17 is shown in FIG. 7A, and the transmission signal at the two-wire line connection terminal 11 is shown in FIGS. 7B and C, respectively. ) has no error, so there is no retransmission, and Di(1-1,...
. . , k) follows Si, the signal S in FIG. ..., N) is sent. In order to reduce the decoding delay, the signal S1 must be transmitted with the smallest possible delay time, but the last signal SN to be transmitted must be sent after the signal SN at the input terminal 17 (Fig. 7A). Because it doesn't happen,
As shown in FIG. 7B, the transmission timing with the minimum transmission delay is to transmit the burst SN immediately after the signal SN in FIG. (1-1, . . . k) continues. Therefore, the transmission delay time is (k+1)Tf
'...(2). On the other hand, as shown in FIG. 7C, there is a possibility that the burst S1 will be erroneous k times, so the final burst S1 will be transmitted after (k+-)Tν...(3)
There may be delays. The total value of equations (2) and (3) (
2k+-)Tv... During (4), the transmission temporary storage device to which the signal S1 is first input is not released. During this time the 7th
Assuming that the signal Si (1-1, . . . p) is input as shown in FIG. A, it can be seen that the minimum number of transmission temporary storage devices is p. However, the following relationship must hold between the time PTf until the signal Sp is input and the equation (4). When this equation (5) is transformed using equation (1), it becomes.

That is, when N and k are given, the number p of temporary transmission storage devices is an integer that satisfies equation (6). The received signal at the two-wire line connection terminal 11 is as shown in FIG. 7D,
The output signal at output terminal 16 is as shown in FIG. 7E.

In order to be able to receive burst R1 after retransmitting burst R1 k times, q must satisfy.

This equation (7) is transformed using equation (1).

Since the decoding delay is expressed as (N-q)Tf, if the maximum value of the integer q that satisfies equation (8) is QITlax, the minimum decoding delay time is ゝ-5! It becomes 11AA'.

In order to explain the required number of reception temporary storage devices, FIG.
6 is a diagram showing the timing when converting to an output signal in No. 6. FIG.

That is, the figure A shows the received signal at the two-wire line connection terminal 11, and the figure B shows the maximum delay state when converting the burst R1 to the output signal at the output terminal 16 when there are three reception temporary storage devices 15. , the end of the signal R1, that is, the right end point c, is burst R1 because there are only three reception temporary storage devices 15.
The beginning of the later third burst R4, ie, FIG. 8A (
7) Must be temporally earlier than point b. That is,
It is sufficient that there is a period Tf during which the signal R1 is sent between the end point a (5b) of the burst R1. However, in the worst case, the start point a' of the frame period exists before the point a in FIG. 8A. , there is a possibility that the signal R1 cannot be sent immediately after the burst R1 in FIG. The relational expression that must be satisfied between Tf and Tf' is.

Using equation (1), we can replace it with the relational expression for N and k, and in the same way, when there are four reception temporary storage devices 15, that is, and in general, when there are x reception temporary storage devices, that is, .

As explained above, when the error correction method of the present invention is used, the real-time performance on the receiving side is maintained compared to when a time-division control bidirectional transmission method that does not use an error correction method is used. “It also has the advantage of being able to perform error correction with simple hardware.

[Brief explanation of drawings]

Is Figure 1 the conventional time division direction? FIG. 2 is a timing diagram showing an example of the operation of FIG. 1. FIG. 3 is a diagram showing the basic principle of the error correction method according to the present invention. FIG. 4 is a diagram showing the basic principle of the error correction method according to the present invention. 5 is a block diagram showing an embodiment of the error correction method according to the present invention. FIG. 6 is a timing diagram showing an example of the operation of FIG. 5. Figure 7 shows the number of transmission temporary storage devices, transmission timing, reception timing, and
FIG. 8, which is a diagram showing the relationship between decoding delay times, is a diagram for explaining the number of reception temporary storage devices for realizing the error correction method according to the present invention. 11: Connection terminal with 2-wire line, 12: Balanced-unbalanced conversion transformer, 13: Transmitter, 14: Receiver, 15: Reception temporary storage device, 17: 4-wire side transmission information input terminal, 16:
4-wire reception signal output terminal, 18: Transmission temporary storage device, 21
: Information bit, 22: Error detection bit, 23: Retransmission request bit, 24, 41: Write address designation counter, 26, 28: Read address designation counter, 29: Read address designation count prohibition circuit, 32: Error detection code 33: retransmission request detection circuit, 34: retransmission request signal input circuit, 35: retransmission request signal generation circuit, 38: retransmission request signal separation circuit, 37: error detection decoder, 39: write address designation count prohibition circuit.

Claims (1)

[Claims] 1. In a time-division direction controlled bidirectional transmission system that transmits and receives 4-wire digital information via a 2-wire line, one or A plurality of dummy bursts are arranged, a plurality of transmission temporary storage devices for waiting for retransmission are provided on the transmitting side, an error detection code generation circuit is supplied with signals read from the transmission temporary storage device, and retransmission from the receiving section is provided. A circuit for detecting a request signal, a circuit for inserting a retransmission request signal into the output of an error detection code generation circuit based on the detected output, and a retransmission control circuit using a dummy burst are provided, and multiple retransmission queues are provided on the receiving side. Error correction for two-wire time-division bidirectional transmission, characterized by comprising a reception temporary storage device, an error detection circuit, and a retransmission control circuit, and inserting an error detection code and a retransmission request signal into a transmission burst. method.