JP3048405B2 - Speech encoder control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control system for a voice encoder in a multiplex transmission apparatus applied to an ATM (Asynchronous Transfer Mode) network.

[0002]

2. Description of the Related Art Conventionally, audio CODEC coding techniques include μ-PCM, ADP
Many coding methods such as CM have been put to practical use. However, these encoding methods have redundancy because information is generated at a constant encoding rate regardless of the nature of the signal. On the other hand, there is an increasing demand for a variable-rate audio CODEC that changes the coding rate in accordance with the magnitude or change of a signal and suppresses deterioration of quality due to coding. This variable-rate audio CODEC encodes only significant information of audio, and the generation of information is asynchronous because the encoding rate is changed. From this, an ATM network capable of processing various bit rates in a unified manner as an application area of the audio CODEC is considered to be promising.

[0003]

When variable rate coding is applied to an ATM network, which is a core technology of broadband ISDN (B-ISDN), it is possible to make the most of its features.
However, when performing voice transmission using an ATM network,
There is a problem that the cell is discarded due to the congestion, and the effect is particularly large in the personal computer communication using the ATM network.

In an ATM network, a waiting time (delay) occurs due to a line, so that echo canceller processing is required. If a signal is received from a modem during the echo canceller processing, the echo canceller processing malfunctions. There is a problem that causes.

The present invention has been made in view of such circumstances, and by controlling the configuration of cells to be transmitted to an ATM network and the operation of an echo canceller according to the nature of an input signal, high quality in an ATM network can be achieved. An object of the present invention is to provide a control method of a voice encoder which enables voice transmission.

[0006]

SUMMARY OF THE INVENTION The present invention relates to a control system for a voice coder that includes an echo canceller, converts an input signal into cells in an ATM format, and sends the cell to an ATM network, as shown in FIG. Detecting means 101 for detecting the property, coding means 102 for coding the input signal at a rate corresponding to the property, cell assembling means 103 for converting the coded signal into cells and transmitting the signal, Correspondingly, the present invention provides a control method for a speech encoder, comprising a control means 105 for stopping the operation of the echo canceller 104.

[0007] The cell assembling means further comprises detection means for discriminating a modem signal output from the modem and its type from the input signal, and sound / silence detection means for detecting sound or silence of the input signal. It is desirable to perform silence compression on a specific type of the determined modem signal.

Further, when the modem detection means detects the modem signal, it is preferable that the modem signal detection state be maintained unless the input signal generates silence for a predetermined time or more.

Further, when the input signal is a facsimile transmission signal comprising a call / terminal identification signal and an image signal, the image signal is encoded and transmitted at a transmission rate higher than the transmission rate of the call / terminal identification signal. Is preferred. The facsimile transmission signal is CCIT
The signal is group 3 according to the classification of TSG14, that is, a G3 facsimile signal.

Further, it is preferable that the control means stops the operation of the echo canceller when the detection means monitors the input signal and detects a signal from the modem.

[0011]

The detecting means 101 detects the property of the input signal, and the coding means 102 codes the input signal at a rate corresponding to the property. Then, the cell assembling means 103 converts the encoded signal into cells and transmits the cells. Therefore, the coding rate of the input signal is switched according to its nature, so that the transmission efficiency is improved. Further, since the control unit 105 stops the operation of the echo canceller according to the property of the input signal, a malfunction of the echo canceller process is eliminated, and transmission of high-quality data is realized.

The transmission efficiency is further improved if the cell assembling means further comprises a modem detecting means and a voiced / silent detecting means, and the cell assembling means performs silence compression on a specific type of the detected modem signal. Also, the input signal is a facsimile signal composed of a calling / terminal identification signal and an image signal, and the image signal is encoded and cellized at a transmission rate higher than the transmission rate of the calling / terminal identification signal, thereby achieving high quality. Can be transmitted efficiently. Further, once the modem detection means detects the modem signal, the modem signal is transmitted preferentially by maintaining the detection state of the modem signal unless the input signal generates silence for a predetermined time (for example, 3 seconds) or more. Is done.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail based on embodiments shown in the drawings. This does not limit the present invention.

FIG. 2 is a block diagram showing an embodiment of the present invention. In FIG. 2, reference numeral 1 denotes a 32 kbps ADPCM (adaptive differential PCM) coding unit, which compresses information of 64 kbps μ-PCM coded data input via the echo canceller 8 into half. Reference numeral 2 denotes a 24 kbps ADPCM encoding unit, which compresses the input μ-PCM encoded data to 1/3. Reference numeral 3 denotes a sound / non-speech detecting unit for judging the presence or absence of sound from the input μ-PCM encoded data.

Reference numerals 5 and 6 denote modem detection sections for monitoring input signals and detecting signals output from the modem. The modem detection section 5 is defined in CCITT Recommendations V.21, V.22 and V.27. The modem detector 6 detects an output signal from a modem (hereinafter, referred to as a low-speed modem), and detects an output signal from a modem (hereinafter, referred to as a high-speed modem) defined in CCITT Recommendation V.29.

Reference numeral 7 denotes an echo canceller control unit. When the modem detection unit 5 or 6 detects an output signal from the modem (hereinafter referred to as a modem signal), the operation of the echo canceller 8 is stopped (disabled). ).

Reference numeral 4 denotes a cell assembling unit, which receives an input signal (μ-P
CM encoded signals), outputs from the encoding units 1 and 2, the sound / silence detection unit 3 and the detection units 5 and 6, and performs cell processing and silence compression processing of the encoded data, According to the cell discard priority determined according to the nature of the signal, the AT
Send cells to the M network. Reference numeral 9 denotes a cell decomposing unit that decomposes cells received from the ATM network. Reference numeral 10 denotes a decryption unit, which decrypts the decomposed received data in the cell decomposition unit 9.

The operation in such a configuration will be described with reference to the flowchart shown in FIG. 3 and the time charts shown in FIGS.

First, in step 101 of FIG. 3, when the detection unit 5 or 6 does not detect a modem signal from an input signal, it is determined that the input signal is a voice signal.
The cell is formed in the voice mode and transmitted to the ATM network (step 107). That is, in this audio mode, a signal coded at 64 kbps μ-PCM is coded at 32 kbps or 24 kbps ADPCM in the coding unit 1 or 2,
The cell assembling unit 4 converts the signal into a cell. At this time, the cell discarding priority is set high, that is, set to “1”, and silence compression processing is performed. In this audio mode, the echo canceller control unit 7
Maintain the echo canceller 8 in the operating state (ON).

Next, in step 101, the detecting unit 5
When detecting the low-speed modem signal output from the V.21, V.22, or V.27 modem, the echo canceller control unit 7 stops (OFF) the operation of the echo canceller 8 (step 1).
02) If the detection unit 3 determines that the input signal is not silent for more than 3 seconds (step 103), it is determined whether the image signal is a facsimile transmission signal image signal or a high-speed modem signal from a V.29 modem. (Step 104), otherwise the low speed modem mode is executed (step 105).

This operation is shown in the time chart of FIG. 4. An input signal coded at 64 kbps μ-PCM is coded at 32 kbps ADPCM in the coding unit 1 and input to the cell assembling unit 4. . At this time, the cell discard priority (Ce
ll-Loss-Priorty) is low, that is, set to 0, and the cell assembling unit 4 executes the silent compression processing. If the input signal is silent for more than 3 seconds (step
103), it is detected by the detection unit 3, and the mode is switched to the voice mode again (step 107).

Next, in step 104, the input signal is
If the modem is a high-speed modem from a V.29 modem, the high-speed modem mode is executed (step 106). In this mode, the signal processed in the above-described voice mode (32 kbps ADPCM) is coded in 64 kbps μ-PCM, the discard priority is set to low, that is, set to 0, and the cell assembling unit is not compressed without silence. At 4, it is cellized. And
In this mode as well, if the silent state of the input signal continues for 3 seconds or more (step 103), it is detected by the detection unit 3 and the mode is switched to the audio mode (step 107). This situation is shown in the time chart of FIG.

When the input signal is a G3 facsimile transmission signal, the facsimile transmission signal is a call signal (1100 Hz intermittent tone) or a called terminal identification signal (2100
(Hz continuous tone), it is detected by the detector 5 as a low-speed modem signal, and the low-speed modem mode (step 105) is executed. Thereafter, the image signal of the facsimile transmission signal is detected by the detection unit 6 as a high-speed modem signal (step 104), and the high-speed modem mode is executed (step 106). If the input signal is silent for three seconds or more (step 103), the detection is detected by the detection unit 3, and the mode returns to the audio mode (step 107). This situation is shown in the time chart of FIG.

The encoding unit 2 shown in FIG.
It converts the input signal coded by PCM into a signal of 24 kbps ADPCM. When the change amount of the input signal is small (for example, when changing from a silent section to a sound section, or from a sound section to a silent section) At the time of change).

[0025]

According to the present invention, since the coding rate is switched according to the nature of the input signal, the transmission efficiency is improved. In addition, when the output signal from the modem is detected, the operation of the echo canceller is stopped, and the facsimile transmission signal is divided into a high-speed signal and a low-speed signal. It can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the principle of the present invention.

FIG. 2 is a block diagram showing an embodiment of the present invention.

FIG. 3 is a flowchart illustrating the operation of the embodiment of the present invention.

FIG. 4 is a time chart showing the operation timing of the embodiment of the present invention.

FIG. 5 is a time chart showing the operation timing of the embodiment of the present invention.

FIG. 6 is a time chart showing the operation timing of the embodiment of the present invention.

[Description of Codes] 1 32k ADPCM coding unit 2 24k ADPCM coding unit 3 voiced / silent detection unit 4 cell assembling unit 5 modem detection unit 6 modem detection unit 7 echo canceller control unit 8 echo canceller 9 cell decomposition unit 10 code Transformation part

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04L 12/28 H04L 12/56 H04B 3/23 H04N 1/00

Claims (4)

(57) [Claims] An echo canceller is provided for receiving an input signal.
In the control method of the speech coder to be sent to the ATM network and cell of the ATM format, a detection means to detect the nature of the input signal, and reference numeral catheter stage to encode at a rate corresponding to the input signal to the nature a cell assembly means to delivery to the cells of the encoded signal,
And a control means for stopping the operation of the echo canceller over in response to the nature of the input signal, a modem signal output from the modem
Modem detection method to determine the signal and its type from the input signal
Step and sound / silence detection to detect sound or silence of the input signal
Output means, and a modem signal for which the cell assembling means is determined.
Speech encoder control method for performing silence suppression for a specific type of issue. 2. When the modem detecting means detects a modem signal, unless the input signal generates silence for a predetermined time or more,
Control Method of speech coder of claim 1, wherein the modem signal detection state is maintained. 3. When the input signal is a facsimile transmission signal comprising a call / terminal identification signal and an image signal, the image signal is encoded and transmitted at a transmission rate higher than the transmission rate of the call / terminal identification signal. The control method for a speech encoder according to claim 1, wherein the encoding is performed. 4. The control method according to claim 1, wherein the operation of the echo canceller is stopped when the detecting means detects the modem signal output from the modem.