JPS62199146A - Data transmission equipment - Google Patents

Abstract

PURPOSE:To resend the error of a data requested by data transmission in real time, to decrease the packet abolition rate and to send the data with high quality by providing a transmission control section, a reception control section and a reception buffer memory. CONSTITUTION:A line interface 8 at the reception side of a data transmission equipment receives a data from a transmission line 7 and a transmission control section 9 processes the ring history and the protocol of a network layer. A packet header in the data from the control section 9 is removed by a packet decomposition section 10, only the packet data is extracted and sent to a reception buffer memory 11. Further, the reception data stored in the memory 11 is read in a voice reproducing section 13, subjected to D/A conversion to form an audible tone. Then a reception control section 12 reads the sequence number of the packet stored in the memory 11, detects an error in the reception data, instructs data read to a recovery section 13 to send a data with high quality.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention (Field of Industrial Application) The present invention relates to a data transmission device having an error control function.

(Prior art) Conventionally, when transmitting voice data via a packet network, real-time performance is required, so even if there is a transmission error, there is no retransmission, and data with errors are simply discarded. I was taking it.

This situation will be explained with reference to FIG.

In FIG. 5, al indicates the generation timing of audio data on the transmitting side, and bl indicates the arrival timing of the audio data on the receiving side. Further, b2 indicates the reproduction timing of audio data on the receiving side.

Further, it is assumed that the generation rate of audio data on the transmitting side is 64 kbps, and the speed of the transmission path is 10 Mbps. Furthermore, it is assumed that the packet size is 151,113 minutes of audio data (120 bytes) that is transmitted as one unit.

An example of the format for sending this data is shown in FIG. Audio data is stored in packet data 307,
Packet header 30 for this packet data 307
6, additional bit 305 called Fe2 to realize error control, and F bit 30 for frame synchronization.
1, an A bit 302 indicating address information, a C bit 303 for controlling frame data, etc. are added and transmitted.

At this time, as shown in Figure 5, 15 m5 (120
Each time a byte of audio data is generated, packet data is sent to the receiving side in the frame structure described above.

For example, packet v(1) arrives at the receiving side after a certain transmission delay and is temporarily stored in a buffer memory (not shown).

Then, after waiting for a predetermined time T, the data is taken out from the buffer memory and played back. The meaning of starting playback after a predetermined time T is that the time it takes for a packet to arrive varies depending on the usage status of the transmission path and the processing status of the exchange and transmitting/receiving terminal, and this variation is absorbed to smooth the received data. This is for outputting to.

In this way, packet v(1) is reproduced, and since packet v(2) has also arrived, it is continuously reproduced.

In this way, packets V(3) and V(4) are also played back continuously, but packet V(5) is discarded due to a transmission error, so although packet V(6) has already arrived, packet V( 5) Silent data is forcibly inserted. After this silent reproduction, packet V6 is reproduced.

Conventionally, such processing was performed, and the reproduction was performed by jumping from packet V (4) to packet - (6), resulting in a deterioration in sound quality.

In order to prevent such packet discards, retransmission processing of error data using a well-known error control procedure may be introduced.

However, when real-time performance is required, such as voice packet communication, if too much time is required for retransmission, the voice quality may improve, but there is a drawback in that smooth conversation will not be possible.

For this reason, in the past, emphasis was placed on real-time performance, and error retransmissions were not performed at the expense of sound quality.

(Problems to be Solved by the Invention) As described above, conventional data transmission devices have a problem in that they cannot perform data transmission in real time without degrading data quality in response to transmission errors.

The present invention has been made in view of this problem, and it is an object of the present invention to provide a data transmission device that can perform data transmission in real time with less deterioration of data quality.

[Structure of the Invention] (Means for Solving the Problems) In order to solve the above-mentioned problems, the present invention provides a data transmission method for transmitting packetized data between other data transmission devices via a transmission path. In the transmission device, a buffer memory temporarily stores received data, a reproduction processing section that reproduces the data stored in the buffer memory, an error detection means that detects errors in the received data, and a If an error is detected in a packet, the first time required for the reproduction processing unit to reproduce the data stored in the buffer memory, and a retransmission request for the packet in which the error was detected, is transmitted again. and a second time required for the packet to be retransmitted and received via the packet, and if the first time is longer, a request is made to retransmit the packet in which an error has been detected. .

(Function) Therefore, according to the present invention, when a data transmission error occurs, retransmission is executed if sufficient data is accumulated in the data buffer, and data transmission is performed in real time with less deterioration of data quality. be able to.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

1 and 2 are block diagrams of one embodiment of the present invention. Here, an example of so-called voice packet communication will be explained in which voice is converted into digital information through A/D conversion and is further transmitted to the other party through packet transmission.

FIG. 2 is a block diagram of the transmitter in this embodiment.

First, the voice generating section 1 corresponds to a transmitter/receiver of a telephone, and has the function of converting a human voice into an electrical signal and converting it into digital information using an A/D converter.

Normally, a sampling frequency of 8K)lz is used, and the data is quantized to 8 bits using a predetermined encoding algorithm.

Next, the digital audio data obtained by the audio generator 1 is passed to the packet assembler 2, which forms packet data 307 for each predetermined bit length as shown in FIG. assembly is performed. Note that at this time, the contents of the packet data are checked and only data containing audio is processed as a packet to be transmitted, but silent data is discarded and not transmitted.

A sequence number is assigned to the packet header 306, including the discarded silent packets, according to the order in which the packets were generated. Therefore, a number omission occurs in the sequence number of the transmitted packet.

The transmission packets formed by the packet assembling section 2 are sequentially sent to the transmission buffer memory 3 and stored therein.

The transmission control unit 4 is 0 proposed by l5O (International Organization for Standardization).
3I (Open System Interconnection) link layer and network layer protocols. In this embodiment, X.25, which is recommended by CCITT (Consultative Committee for International Telegraph and Telephone), is used as the protocol. Communication progresses while exchanging data with the other terminal based on this x.25 procedure. The data to be transmitted is sent to the transmission line 7 via the line interface 6 corresponding to the physical layer of O8I under the control of the transmission control unit 4.

On the other hand, FIG. 1 is a block diagram of the receiving section in this embodiment.

Data received via the transmission path 7 is input to the transmission control section 9 via the line interface 8.

This transmission control section 9 also has a function of processing X, 25 procedures corresponding to the transmission control section 4 of FIG.

The packet decomposition unit 10 removes the packet header 306 from the received data sent from the transmission control unit 9, extracts only the packet data 307, and stores it in the reception buffer memory 1.
Transfer to 1.

The audio reproduction unit 13 reads out the received data stored in the reception buffer memory 11, performs D/A conversion, and reproduces it into audible sound.

The reception control unit 12 reads the sequence number of the packet stored in the reception buffer memory 11 from the packet decomposition unit 10, and instructs the audio reproduction unit to insert silent data into the silent portion. In addition, in the case of audio packet playback, 1. An instruction is given to read data from the reception buffer memory 11.

Furthermore, this reception control unit 12 has a function of detecting errors in received data, a function of measuring the amount of data stored in the buffer 1 memory when an error is detected in the received packet, and a function of measuring the amount of data stored in the buffer 1 memory. The function calculates the first time required until the data stored in the memory is reproduced by the reproduction processing unit, and requests retransmission of the packet in which an error has been detected, so that the packet is transmitted again via the transmission path. A function that calculates a second time required for a packet to be retransmitted and received, and a function that compares the first time and the second time, and if the first time is larger, a packet in which an error has been detected. If the second time is longer than the second time, it has the function of issuing an instruction to discard the packet in which an error has been detected.

Next, data exchange in this embodiment will be explained in detail.

FIG. 4 is a diagram showing an example of data exchange in this embodiment. It is assumed that a transmission error occurs in packet (5) of the transmitted data, similar to the conventional example shown in FIG.

When a transmission error is detected on the receiving side, in the conventional process, the next packet V(6) is received as a discard process, but in this embodiment, the reception control unit 12 controls the amount of data accumulated in the reception buffer memory 11. Discern.

At this time, packet ■(2) is being played back by the audio playback unit 13, and packet ■(2) is already stored in the reception buffer memory 11.
3) and v(4) have already arrived.

Therefore, the reception control unit 12 calculates that it will take at least 15m5×2 time until the reproduction of these two packets V(3) and V(4) is completed.

On the other hand, the time from when a retransmission is requested until the retransmitted data arrives depends on the transmission speed of the transmission path, the data length, the processing speed of the terminal or processing device on the transmission path, such as exchange Ia, and the traffic condition. be.

Now, if the packet size length is 15m5 and the data generation rate is 64kbps, then 15m5 will be 120 bytes. Further, assuming that the transmission speed of the transmission line is 10 Mbps, 96 μs is required to transmit 120 bytes. In addition, although the other processing time mentioned above requires several ms, the packet V(3
), the reception control unit 12 determines that there is a sufficient possibility of retransmission within the playback time (30 m5) of V(4).

Therefore, in FIG. 4, the receiving side requests retransmission of packet 2 (5), but the transmission of packet 2 (6) has already started, and packet V (6) will be received first. However, this packet is discarded.

On the other hand, upon receiving the retransmission request, the transmitting side retransmits the packet data of packet V(5) stored in the transmitting buffer memory 3 after transmitting packet (6). Subsequently, packet v(6) is retransmitted, and then packet ■(7) already stored in the transmission buffer memory 3 is retransmitted.
is also sent. When packet (7) is transmitted, the transmission buffer memory 3 becomes empty, and no transmission is performed until packet (8) occurs.

On the other hand, when the receiving side discards packet (6) and waits for the next reception, packet (5) is retransmitted. This time, it is stored in the transmission buffer memory 11.

At this time, the reproduction of packet (4) has just started, and there is enough time for the reproduction of packet V (5).

Therefore, if the retransmitted packet (5) is played after the reproduction of the packet (4) is completed, continuous audio can be reproduced without packet loss.

Note that after receiving packet ■(5), packet V
(6) and V(7) are received, but since these packet data are temporarily stored in the reception buffer memory, there is no harm caused by retransmission of packet (5).

If the time required to retransmit packet V(5) is longer than the time required for the audio reproduction section 13 to reproduce the data stored in the reception buffer memory 11, the reception control section 12 If so, the reception control unit issues an instruction to discard the erroneous packet V(5).

As described in detail above, in conventional voice packet communication, packets are discarded even though there is time because retransmission is not performed, causing deterioration of the reproduced sound. In other words, retransmission occurs as long as there is time to spare, and the packet discard rate can be reduced.

Therefore, the reproduced sound quality can be improved and voice packet communication can be effectively realized.The present invention can be modified in various ways within its technical concept.

That is, although the case of voice packet communication has been described in this embodiment, it is clear that the present invention can also be applied to other data transfers that require real-time performance.

Furthermore, although the explanation has been given using CCITT Recommendation X and 25 as an example of the transmission control procedure, the present invention is also applicable to other procedures having an error retransmission function.

Furthermore, regarding the link layer and below in the O3I model, whether it is the C3MA/CD (carrier detection multiple access collision detection) method or other methods, the present invention is independent of these methods and does not apply to these methods. It is also clear that it does not depend on

[Effects of the Invention] As explained in detail above, even in data transmission that requires real-time performance, error retransmission is performed as much as possible, so the packet discard rate can be reduced and data with better quality can be obtained. transmission can be realized.

[Brief explanation of drawings]

FIG. 1 is a block diagram of the receiving side in one embodiment of the present invention, FIG. 2 is a block diagram of the transmitting side in the same embodiment, and FIG.
FIG. 4 is a diagram showing the structure of a frame sent to a transmission path, FIG. 4 is a diagram showing data exchange in the same embodiment, and FIG. 5 is a diagram showing a conventional data transmission system. 7...Transmission line 11...Reception buffer memory 12...
......Reception control unit 13...Audio reproduction product applicant Toshiba Corporation Patent attorney Sasa Suyama - Fig. 1 Fig. 2 Fig. 3 Fig. 4 Fig. 5

Claims (1)

[Claims] In a data transmission device that transmits packetized data between other data transmission devices via a transmission path,
A buffer memory for temporarily storing received data, a reproduction processing unit for reproducing the data stored in the buffer memory, an error detection means for detecting errors in the received data, and a reproducing unit for detecting errors in the received packets. If an error is detected, a request is made to retransmit the packet in which the error was detected and the first time necessary for the reproduction processing unit to reproduce the data stored in the buffer memory, and the packet is transmitted again through the transmission path. and a second time required for the packet to be retransmitted and received, and if the first time is longer, a request is made to retransmit the packet in which an error has been detected. Transmission device.