JPS5840399B2 - Voice packet call setting method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a voice packet call setting method that prevents deterioration in call quality in voice packet communications.

Audio data obtained by digitizing audio signals is separated into predetermined data lengths, and receiver address information is added to each piece of data to form audio packets, and this audio packet is sent to a transmission path to create a packet. 2. Description of the Related Art Voice packet communication is known in which voice packet communication is transmitted to a receiving side via an exchange or the like.

Since audio packets are sent out at time intervals, the receiving side temporarily stores the audio packets in a buffer memory and reproduces the audio signal through decoding processing.

However, the transmission delay time of audio packets varies depending on the packetization process and the state of the transmission system. Therefore, a method is proposed in which the buffer memory capacity on the receiving side is specified in advance and received audio packets are stored and then sequentially decoded. However, it is difficult to keep the transmission time below the allowable value, resulting in poor call quality.Also, with the method of changing the decoding timing by adaptive control, the configuration becomes complicated and stable call quality cannot be obtained. There were drawbacks.

An object of the present invention is to add time data to voice packets, compensate for fluctuations in transmission delay time based on this time data, and obtain a constant call quality.

Examples will be described in detail below.

FIG. 1 is a block diagram of an embodiment of the present invention, in which a voice signal from a telephone 1 is converted into a PCM signal etc. in a transmitter 2 and packetized, and time data from a time data circuit 3 is The audio packet is added to the transmission line 5 and sent out to the transmission path 5.

Also, the audio packets received via the transmission path 5 are sent to the receiving section 4.
At this point, the time data is separated and decoded based on this time data, and the voice signal is sent to the telephone set 1.

Since time data is added to the voice packet in this way, it is possible to compensate for fluctuations in transmission delay time and always decode with a constant delay time, thereby stabilizing call quality.

FIG. 2 is a more detailed block diagram of an embodiment of the present invention, in which 11 is a telephone, 12 is an encoder, 13° 18, 22,
23 is a buffer, 14 and 24 are packet serial number counters,
15 is a control unit, 16 is a memory, 17 is a time data storage circuit, 19 and 21 are high level data link control circuits, 2
0 is a transmission path, and 25 is a decoder.

Although this embodiment is for the case where there is only one telephone 11, it can also be applied to the case where multiple telephones are connected by simply adding a dial number determination circuit, and it can also be used as a high-speed line interface circuit. Although the level data link control circuits 19 and 21 are used, it is also possible to use line interface circuits with other configurations.

A voice signal from the telephone 11 is converted into, for example, a PCM signal by an encoder 12 and stored in a buffer 13.

When a predetermined number of sample data of the audio signal, that is, audio data blocks are accumulated in this buffer 13, the encoder 12 instructs the packet serial number counter 14 to count up, and sends an audio data block transfer instruction to the control unit 15. do.

The control unit 15 thereby transfers data from the buffer 13 to the memory 1.
6, the packet serial number is transferred from the packet serial number counter 14, and the time data storage circuit 1
The time data is read from each of the buffers 18 and 7 and written in such a way as to be added to the audio data block written to the memory 16.
The audio data block is packetized via the level data link control circuit 19, and the audio packet is transferred to the transmission path 2.
Send to 0.

The time data storage circuit 17 has a clock function synchronized with the packet transmission system and outputs time data, and the output of a counter can also be used as the time data.

When a voice packet with time data added is received via the transmission path 20, the high-level data link control circuit 2
1, the packet is disassembled, data portions such as audio data blocks and time data are stored in the buffer 22, and a transfer instruction request is sent to the control unit 15.

The control unit 15 thereby controls the transfer of the contents of the buffer 22 to the memory 16, and transfers the contents of the buffer 22 to the time data storage circuit 17.
The transmission delay time is determined from the time data read from the memory 16 and the time data in the received audio packet in the memory 16.

This transmission delay time is compared with a predefined allowable transmission delay time, and if the transmission delay time is larger than the allowable transmission delay time, the audio data block is discarded, and if it is within the allowable transmission delay time, it is allowed. After the audio data block is delayed by the difference between the transmission delay time and the transmission delay time in the memory 16, it is transferred to the buffer 23, decoded by the decoder 25 to become an audio signal, and the audio signal is added to the telephone 11. The decoder 25 also instructs the packet serial number counter 24 to count up.

As explained above, the present invention provides the time data storage circuit 1
7. Time data is added to the audio packet by the time data adding means consisting of the control unit 15 and the like, and the time data is sent out, and the time data in the received audio packet and the time of reception are added by the delay amount adjusting means by controlling the reading timing of the memory 16, etc. The transmission delay amount of received audio packets is controlled to be constant according to the difference from the time data of This has the advantage of being constant within the delay time and providing stable call quality.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of the invention, and FIG. 2 is a detailed block diagram of an embodiment of the invention. 1.11 is a telephone, 2 is a transmitter, 3 is a time data circuit,
4 is a receiving section, 5 and 20 are transmitting sections, 12 is an encoder, 13,
18, 22, and 23 are buffers, 14.24 is a packet serial number counter, 15 is a control unit, 16 is a memory, 17 is a time data storage circuit, 19.21 is a high-level data link control circuit, and 25 is a decoder.

Claims (1)

[Claims] 1. A transmitting side is provided with a time data adding means, and the adding means adds time data to an audio packet and sent. A receiving side is provided with a delay amount adjusting means, and the delay amount adjusting means adjusts each received audio packet. A voice packet call setting method characterized in that control is performed according to the difference between time data in each received voice packet and time data at the time of reception so that the amount of transmission delay is constant.