JP3075246B2 - Voice packet transmitting / receiving method and apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for transmitting and receiving a voice packet, which encodes and compresses a voice analog signal, transmits a packet in a sound section, and absorbs network delay fluctuations on a receiving side with a buffer.

[0002]

2. Description of the Related Art Conventionally, this type of voice packet transmitting / receiving apparatus has been disclosed in, for example, Japanese Patent Application Laid-Open No. 4-188929, in audio packet communication, performing reproduction by absorbing delay fluctuation of voice packets. A voice packet control device aiming at reducing the packet discard rate has been applied.

FIG. 8 shows a configuration of a voice packet transmitting / receiving apparatus to which the voice packet control apparatus is applied, and FIG.
FIG. 10 shows a configuration of a packet disassembly control unit 200 shown in FIG. 8, and FIG. 11 shows a format of a voice packet used in the voice packet transmitting / receiving apparatus shown in FIG. Is shown.

The voice packet communication system shown in FIG. 9 comprises a voice packet transmitting / receiving apparatus 2, a packet relay node 3, and relay lines A1 to A3. The voice packet transmitting / receiving device 2 has a transmitting function of encoding and compressing a voice analog signal and then packetizing the voice packet, and a receiving function of outputting a voice analog signal after decoding and expanding a voice packet. The packet relay node 3 is installed at each connection point of the relay lines A1 to A3 and relays a packet.

[0005] The voice packet transmitting / receiving apparatus 2 has a configuration shown in FIG. The A / D converter 601 converts an audio analog signal into a digital signal. Encoding compression section 600
Performs encoding and compression of the audio digital signal. The sound / silence detection unit 500 detects a sound / silence from the audio digital signal.
Determine silence. The packet assembling control unit 400 assembles the audio digital signal into a packet, and
Sends voice packets to the network 1 and receives voice packets from the network 1. The packet disassembly control unit 200 disassembles the received voice packet and absorbs the delay fluctuation by the fluctuation absorbing buffer. The decoding / decompression section 300 decodes / decompresses the audio data decomposed by the packet decomposition control section 200, and the D / A conversion section 301 converts the audio data into an analog signal and outputs the analog signal.

The packet disassembly control section 200 has an internal configuration shown in FIG. The header separating unit 202 separates a header portion from the audio packet. The fluctuation absorbing buffer 205 is a buffer that stores the audio data separated by the header separating unit 202. Congestion state detector 212
Then, the congestion state is detected from the header information, and the additional fixed delay control unit 211 determines an additional fixed delay for absorbing fluctuations according to the congestion state. The packet reproduction control unit 210 controls the audio data stored in the fluctuation absorbing buffer 205.

Next, the operation will be described. The audio analog signal is converted into a digital signal by the A / D converter 601 and is coded and compressed by the coder / compressor 600. The output of the A / D converter 601 is output to the sound / silence detector 5.
00 determines whether there is sound or no sound, and the information is passed to the packet assembly control unit 400. Packet assembly control unit 40
In the case of 0, the encoded and compressed audio data is formed into an audio packet only for a sound part based on the sound / non-sound information, passed to the communication control unit 700, and transmitted to the network 1. here,
In the header of the voice packet, a congestion indication bit 31 is provided in a layer 3 header 30 as shown in FIG. The voice packet transmitted to the network 1 is shown in FIG.
The packet relay node 3 receives the packet and distributes it for each destination. If congestion occurs, the packet relay node 3 performs queuing within the allowable delay range and discards the packet exceeding the allowable range. Here, the congestion indication bit 31 of the queued voice packet includes a congestion indication = 1.
Is set.

[0008] The voice packet received by the communication control unit 700 is passed to the packet decomposition control unit 200, the header is separated by the header separation unit 202 in FIG. 10, and the voice data is accumulated in the fluctuation absorbing buffer 205.

On the basis of the separated header information, the congestion state detecting section 212 judges that the packet is in the congestion state when a certain amount of packets with the congestion indication bit 31 of the received voice packet turned on are received. Is notified to the additional fixed delay control unit 211. The additional fixed delay control unit 211 determines the additional fixed delay based on this information, and notifies the packet reproduction control unit 210. In the packet reproduction control unit 210,
The packet is accumulated in the fluctuation absorbing buffer 205 for a time corresponding to the additional fixed delay determined for the first packet in the sound section, and then transmitted.

[0010]

When the above-mentioned conventional voice packet transmitting / receiving apparatus is applied to an existing network, the congestion indication bit of the layer 3 header of the voice packet is detected by detecting congestion at a relay node of the network. There is a drawback that the method cannot be applied as it is without adding a process for operating.

An object of the present invention is to provide a voice packet transmitting / receiving method and apparatus which reduce the packet discard rate by absorbing the delay fluctuation and reproducing the voice without changing the processing contents of the relay node of the network. .

[0012]

A voice packet transmitting / receiving method according to the present invention includes the steps of encoding and compressing an audio analog input signal for performing conversational audio communication via a network; Detecting a voiced / silent state of the signal and packetizing voice data only in the voiced section; transmitting and receiving the voice packet; decomposing the received voice packet into a header portion and a voice data portion; A voice packet transmitting / receiving method, comprising the steps of: absorbing a delay fluctuation of the voice data; and performing a decoding process on the voice data transmitted from the buffer to output a voice analog signal. And the step of absorbing the delay fluctuation of the audio data by the buffer comprises the steps of: Receiving the audio data, the audio data is stored in a buffer in which a threshold is set by a specified amount and transmitted from the buffer. Only when the threshold value is exceeded, the audio data is discarded from the oldest one.

With such a configuration, it is not necessary to change the processing content of the relay node, and the voice data discard rate can be reduced.

The step of decomposing the voice packet into a header part and a voice data part and absorbing a delay fluctuation of the voice data by a buffer includes monitoring a number of times of underflow of the buffer and a number of times of voice data discarding in the buffer. And the step of determining the initial storage amount of the buffer in the next voiced section based on the number of times of underflow and the number of times of discarding audio data.

[0015] The step of determining the initial accumulation amount of the buffer in the next sound section is performed when the number of underflow occurrences per unit time exceeds a set threshold. Increasing the initial accumulation amount when no underflow occurs and no audio data is discarded.

A voice packet transmitting / receiving apparatus according to the present invention comprises an A / D converter for converting a voice analog input signal into a digital signal for performing conversational voice communication via a network, and encoding and compressing the digital signal. A voice / silence detection unit for detecting whether the digital signal is a voiced section or a voiceless section, and voice data of only the voiced section of the coded and compressed signal. A packet assembling control unit for packetizing a packet, a communication control unit for transmitting a voice packet to a network and receiving the voice packet from the network, and disassembling the received voice packet into a header portion and a voice data portion, and delaying the voice data by a buffer. A packet disassembly control unit that absorbs fluctuations, and a decoding decompression unit that decodes and decompresses output voice data of the packet decomposition control unit. In a voice packet transmitting / receiving apparatus comprising: a D / A converter for converting an output digital signal of the decoding / decompression section into a voice analog signal and outputting the voice signal, the packet decomposition control section converts the input voice packet into a header section and voice data. A header separation unit for decomposing the data, storing the audio data, and a fluctuation absorbing buffer in which a threshold value of the stored data amount is set, a timer, and a threshold value in which the stored data amount in the fluctuation absorbing buffer is the set threshold value. A buffer monitoring unit that monitors whether or not the elapsed time has elapsed and measures an elapsed time after the threshold value has been exceeded in cooperation with the timer, and an elapsed time measured by the buffer monitoring unit has exceeded a set time. In this case, a buffer control unit for discarding the audio data in the fluctuation absorbing buffer from the oldest one is provided.

The packet disassembly control unit stores and holds the number of times audio data is discarded in the fluctuation absorbing buffer discarded by the buffer control unit and the number of times underflow occurs in the fluctuation absorbing buffer, and discards the audio data. The present invention further includes an information storage unit for determining an initial storage amount of the next sound section of the fluctuation absorbing buffer based on the number of times and the number of times of underflow occurrence.

When determining the initial storage amount of the buffer in the next sound section, the information storage unit determines whether the number of underflow occurrences per unit time exceeds a set threshold.
The initial storage amount of the next sound section may be increased, and the initial storage amount may be reduced when there is no underflow and no audio data is discarded.

[0019]

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a flowchart of a first embodiment of a voice packet transmitting / receiving method according to the present invention, and FIG. 2 is a format diagram of a voice packet used in the voice packet transmitting / receiving method of FIG.

The voice packet transmitting / receiving method according to the present embodiment comprises:
As shown in the flowchart of FIG. 1, first, an analog audio input signal for performing conversational audio communication via a network is A / D converted and coded and compressed (step S1).

Next, the sound / non-speech state of the encoded and compressed signal is detected, and audio data is packetized only in the sound period (step S2). Here, the audio packet has a header section 10 and an audio data section 11 as shown in the format of FIG.

Subsequently, the voice packet is transmitted and received via the network (step S3).

Subsequently, the received voice packet is separated into a header section 10 and a voice data section 11 (step S4).

Subsequently, when the head voice packet of the voiced section is received, a predetermined amount of voice data is stored in the buffer in which the threshold value is set, and is transmitted from the buffer (step S5).

Subsequently, the audio data is discarded only when the accumulated amount of the audio data in the buffer exceeds the threshold even after a certain period of time has passed since the threshold exceeded (step S6).

Subsequently, the audio data sent from the buffer is decoded and D / A converted to output an audio analog signal (step S7).

In the voice packet transmission / reception method of the present embodiment, voice data is stored in the buffer by a specified amount and transmitted from the buffer, so that the delay fluctuation of the voice packet can be reduced without changing the processing contents of the relay node. Can be absorbed.

Further, even if the audio data stored in the buffer exceeds the threshold value, the audio data is not discarded unless a predetermined period of time elapses, so that discarding of the audio data can be reduced.

FIG. 3 is a flowchart of a voice packet transmitting / receiving method according to a second embodiment of the present invention.

The voice packet transmitting / receiving method according to the present embodiment comprises:
As shown in the flowchart of FIG. 3, steps S11 to S16 are completely the same as steps S1 to S6 of FIG.

Subsequent to step S16, the number of times of underflow in the buffer and the number of discards of audio data in step S16 are monitored and stored (step S17).

Next, based on the number of times of underflow of the buffer and the number of discards of audio data in the buffer, the initial storage amount of the buffer in the next sound section is determined (step S1).
8). Here, for example, if the number of underflow occurrences per unit time exceeds the threshold, the initial storage amount of the next sound section is increased, and if there is no underflow occurrence and audio data is not discarded, the initial storage amount is increased. There is a decision method to reduce

Subsequently, the audio data sent from the buffer is decoded and D / A converted to output an audio analog signal (step S19).

The voice packet transmitting / receiving method according to the present embodiment comprises:
Since the initial storage amount of the next sound section is determined based on the information on the packet discard in the buffer and the number of underflow occurrences, there is an effect of reducing the number of discards of the audio data more than in the first embodiment.

FIG. 4 is a block diagram of one embodiment of a voice packet transmitting / receiving apparatus to which the voice packet transmitting / receiving method of FIG. 1 is applied, and FIG. 5 is a block diagram of the packet disassembly control unit 250 of FIG.

This voice packet transmission / reception apparatus corresponds to 21 in FIG.
As shown in FIG. 8, the packet disassembly control unit 250 replaces the packet disassembly control unit 200 of the voice packet transmitting and receiving apparatus of FIG.
Is provided.

The packet disassembly control unit 250 comprises a header separation unit 202, a buffer control unit 203, a buffer monitoring unit 204, a fluctuation absorbing buffer 205, and a timer 206, as shown in FIG. Header separation unit 202
Separates a voice packet into a header section 10 and a data section 11 shown in FIG. The fluctuation absorption buffer 205 transmits the audio data after accumulating the audio data by the initial accumulation amount 208. The buffer monitoring unit 204 cooperates with the timer 206 to monitor whether the accumulated data amount in the fluctuation absorbing buffer 205 exceeds the threshold 207, and cooperates with the timer 206 to determine the elapsed time after the threshold 207 is exceeded. measure. When the measured elapsed time exceeds a certain time, the buffer control unit 203 discards the audio data in the fluctuation absorbing buffer 205 from the oldest. With the above configuration, the packet disassembly control unit 250 separates the header of the received audio packet from the audio data, absorbs the delay fluctuation, and passes the audio data to the decoding and decompression unit 300. The processing of the other components is the same as in the case of FIG.

Next, the operation of the voice packet transmitting / receiving apparatus will be described. This device uses parameters (for example, sampling rate, sampling size,
A / D using stereo / monaural selection / compression method)
Upon receiving the audio analog signal, the conversion unit 601 converts the signal into a digital signal, and the encoding and compression unit 600 performs encoding and compression. In the sound / silence detecting section 500, the A / D
The voice data output from the conversion unit 601 is determined to be sound / non-sound, and is notified to the packet assembly control unit 400. The packet assembling control unit 400 forms a voice packet only in a voiced section in the format shown in FIG.
Send to

The voice packet arriving from the network 1 is sent to the packet disassembly controller 20 via the communication controller 700.
Passed to 0. Referring to FIG. 5, the audio packet is divided into header 10 and audio data 11 by header separation section 202.
And the audio data 11 is stored in the fluctuation absorbing buffer 20.
5 is stored. When the initial storage amount 208 is stored in the fluctuation absorbing buffer 205, the transmission of the audio data starts.
The audio data is subjected to a decoding process by the decoding decompression unit 300 and the D / A conversion unit 301 in FIG.

The buffer monitoring unit 204 in FIG. 5 monitors whether or not the accumulated amount exceeds the threshold value 207 set in the fluctuation absorbing buffer 205. The timer 206 measures the elapsed time after the accumulated amount in the fluctuation absorbing buffer 205 exceeds the threshold 207, and if the accumulated amount still exceeds the threshold 207 after the elapse of a predetermined time, the fluctuation absorbing buffer 205 Are discarded from old voice data until the initial storage amount 208 is reached.

In the voice packet transmitting / receiving apparatus of this embodiment,
When the delay fluctuation of the voice packet is absorbed and the voice data stored in the buffer exceeds the threshold,
The old data is discarded, but is not discarded after a lapse of a set time, which has the effect of reducing the number of data discards.

FIG. 6 is a block diagram of an embodiment of a voice packet transmitting / receiving apparatus to which the voice packet transmitting / receiving method of FIG. 3 is applied, and FIG. 7 is a block diagram of the packet disassembly control section 260 of FIG.

This voice packet transmitting and receiving apparatus is the same as the one shown in FIG.
As shown in FIG. 2, the configuration is the same as that of the voice packet transmission / reception device of FIG. 4 except that a packet decomposition control unit 260 is used instead of the packet decomposition control unit 250 of FIG.

As shown in FIG. 7, the packet disassembly control unit 260 stores the information storage unit 2 in the packet disassembly control unit 250 shown in FIG.
01 is added. The information storage unit 201 stores information on the number of times audio data is discarded in the fluctuation absorbing buffer 205 discarded by the buffer control unit 203 and the number of times an underflow occurs in the buffer. The information on the number of times audio data is discarded and the number of times underflow occurs is used to determine the initial storage amount of audio data in the next sound section. For example, if the number of underflows per unit time exceeds the threshold value, the initial storage amount of the next sound section is increased, and if there is no underflow and no audio data is discarded, the initial state of the next sound section is increased. Reduce the amount of accumulation.

The voice packet transmitting / receiving apparatus of this embodiment stores the number of times underflow has occurred and the number of times audio data is discarded, and determines the initial storage amount of the buffer in the next voiced section based on this information. The delay fluctuation due to the network of the packet can be absorbed, and the effect of reducing discard of voice data can be obtained.

[0047]

As described above, according to the present invention, a threshold value is set in a fluctuation absorbing buffer, and a packet is discarded after a lapse of a certain period of time after the accumulated amount in the buffer exceeds the threshold value. Has the effect of reducing the voice packet discard rate in the voice packet transmitting and receiving apparatus, and furthermore, based on the voice data discard and underflow information in the fluctuation absorbing buffer, the initial state of the next sound section. By determining the amount of storage, there is an effect that the voice data discard rate is reduced without changing the processing content of the relay node, and good voice communication becomes possible.

[Brief description of the drawings]

FIG. 1 is a flowchart of a first embodiment of a voice packet transmitting / receiving method of the present invention.

FIG. 2 is a format diagram of a voice packet used in the voice packet transmitting / receiving method of FIG. 1;

FIG. 3 is a flowchart of a voice packet transmitting / receiving method according to a second embodiment of the present invention.

FIG. 4 is a block diagram of an embodiment of a voice packet transmitting / receiving apparatus to which the voice packet transmitting / receiving method of FIG. 1 is applied;

FIG. 5 is a block diagram of a packet disassembly control unit 250 shown in FIG. 4;

FIG. 6 is a block diagram of an embodiment of a voice packet transmitting / receiving apparatus to which the voice packet transmitting / receiving method of FIG. 3 is applied;

FIG. 7 is a block diagram of a packet disassembly control unit 260 of FIG. 6;

FIG. 8 is a block diagram of a conventional example of a voice packet transmitting / receiving apparatus.

9 is a block diagram of a voice packet communication system including the voice packet transmitting / receiving apparatus of FIG.

FIG. 10 is a block diagram of the packet disassembly control unit 200 of FIG.

FIG. 11 is a format diagram of a voice packet used in the voice packet transmitting / receiving apparatus of FIG. 8;

[Explanation of symbols]

 1 Network 21, 22, 23 Voice Packet Transceiver 100, 101 Parameter Negotiation Unit 201 Information Storage Unit 202 Header Separation Unit 203 Buffer Control Unit 204 Buffer Monitoring Unit 205 Fluctuation Absorbing Buffer 206 Timer 207 Threshold 208 Initial Storage 209 Encoding Information Monitoring Units 250, 260, 270 Packet decomposition control unit 300 Decoding / decompression unit 301 D / A conversion unit 400, 450 Packet assembly control unit 500 Voice / silence detection unit 600 Coding compression unit 601 A / D conversion unit

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04L 12/56 H04L 12/28 H04Q 11/04

Claims (6)

(57) [Claims] 1. A step of encoding and compressing an audio analog input signal for performing conversational audio communication via a network, and detecting a sound / non-speech state of the encoded and compressed signal to detect said sound. Packetizing voice data only in the section, transmitting and receiving the voice packet, decomposing the received voice packet into a header portion and a voice data portion, and absorbing a delay fluctuation of the voice data by a buffer; Decoding the audio data sent from the buffer and outputting an audio analog signal, comprising: decomposing the audio packet into a header portion and an audio data portion;
The step of absorbing the delay fluctuation of the audio data by the buffer includes, when receiving the first audio packet of the voiced section, accumulating a predetermined amount of the audio data in a buffer in which a threshold is set and transmitting the audio data from the buffer. And a step of discarding audio data from the oldest one only when the amount of data stored in the buffer exceeds the threshold even after a certain period of time has elapsed after the amount of storage in the buffer has exceeded a threshold. 2. The step of decomposing the voice packet into a header part and a voice data part, and absorbing a delay fluctuation of the voice data by a buffer, wherein the number of underflow occurrences in the buffer and the number of voice data discards in the buffer are determined. 2. The voice packet transmitting / receiving method according to claim 1, further comprising the steps of: monitoring and storing; and determining an initial storage amount of the buffer in a next voiced section based on the number of underflows and the number of voice data discards. . 3. The method according to claim 1, wherein the step of determining an initial accumulation amount of the buffer in the next sound section is performed when the number of underflow occurrences per unit time exceeds a set threshold. 3. The voice packet transmission / reception method according to claim 2, wherein the initial storage amount is reduced when an underflow does not occur and voice data is not discarded. 4. An A / D converter for converting a voice analog input signal into a digital signal for performing conversational voice communication via a network; a coding compressor for coding and compressing the digital signal; A voice / silence detection unit for detecting whether the digital signal is a voice section or a voice section, and a packet assembly for packetizing voice data of only the voice section of the coded and compressed signal into voice packets. A control unit, a communication control unit for transmitting a voice packet to a network and receiving the voice packet from the network, and a packet decomposition control for decomposing the received voice packet into a header portion and a voice data portion and absorbing a delay fluctuation of the voice data by a buffer A decoding unit for decoding and expanding output audio data of the packet decomposition control unit; and an output digital unit of the decoding and expansion unit. An audio packet transmitting / receiving apparatus comprising: a D / A conversion unit that converts a signal into an audio analog signal and outputs the audio analog signal. The packet disassembly control unit includes: a header separation unit that disassembles an input audio packet into a header unit and an audio data unit; Storing a voice data, a fluctuation absorbing buffer in which a threshold of the amount of stored data is set, a timer, and monitoring whether the amount of stored data in the fluctuation absorbing buffer exceeds the set threshold, A buffer monitoring unit that measures an elapsed time after the threshold value is exceeded in cooperation with the timer, and when the elapsed time measured by the buffer monitoring unit exceeds a set fixed time, the fluctuation absorption buffer An audio packet transmitting / receiving apparatus, comprising: a buffer control unit that discards audio data from oldest ones. 5. The packet disassembly control unit stores and holds the number of times audio data is discarded in the fluctuation absorption buffer discarded by the buffer control unit and the number of times underflow occurs in the fluctuation absorption buffer. 5. The voice packet transmitting / receiving apparatus according to claim 4, further comprising an information storage unit for determining an initial storage amount of a next sound section of the fluctuation absorbing buffer based on the number of discards and the number of underflows. 6. The information storage section determines the initial storage amount of the buffer in the next sound section, and when the number of underflow occurrences per unit time exceeds a set threshold, the next sound section. 6. The voice packet transmitting / receiving apparatus according to claim 5, wherein the initial accumulation amount of the section is increased, and when no underflow occurs and audio data is not discarded, the initial accumulation amount is decreased.