JP2000049667A - Intra-voice-band signal device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately estimate echoes without being affected by the difference of polarity inversion characteristics by the kind of terminal equipment connected to a terminal side in-band ringer by performing training for optimizing an echo canceler in a communication preparation period after the DC loop formation of a call termination terminal until the polarity inversion of an exchange. SOLUTION: In this intra-voice-band signal device provided with the echo canceler, a timer function is provided inside a central control means 31 and the timer starts an operation as soon as it detects that the terminal equipment 7 is off-hooked by a DC monitoring means 21 and counts time set beforehand, 500 mSec for instance, which is the time required for the training of the echo canceler to put it concretely. When counting is ended and time-out is performed the polarity inversion of a subscriber's line 11 housing the terminal equipment 7 is performed a communication period is started, communication and data transmission are performed and it continues until on-hooking is performed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an in-voice signal apparatus for performing communication using a dedicated line restricted within an audio band, and more particularly to training of an echo canceller in an in-voice signal apparatus for transmitting a billing signal. About the means.

[0002]

2. Description of the Related Art In a general analog public telephone line, a connection from a general subscriber terminal such as a telephone or a facsimile to an exchange is directly connected by two signal lines.
It is possible to transmit a Hz calling signal, a dial pulse due to opening and closing of a DC circuit, a DC current for transmitting an on-hook or off-hook of a transmitter / receiver, and the like. On the other hand, in a dedicated line, the band used for communication is 300 Hz to 3.4.
In many cases, the voice band is limited to the kHz voice band, and it is impossible to transmit the above-described dial pulse signal, DC signal, or the like as it is. Therefore, when a dedicated line is used, a dial pulse signal, an on-hook signal, an off-hook signal, or the like is temporarily converted to a signal in a voice band,
It is necessary to inversely convert the converted in-band signal into the original dial pulse signal, on-hook signal, off-hook signal, or the like. A device having such a function is referred to as an "in-band signal device (hereinafter, in-band ringer)", and the above-described mutual conversion is performed by being provided between dedicated communication lines. This in-band linger is
By having a function of superimposing a predetermined control signal within the communication band, various signals such as the above-mentioned 16 Hz calling signal, a dial pulse signal by opening and closing of a DC circuit, and a hook signal for transmitting an on-hook or off-hook of a transceiver are transmitted. Is temporarily replaced with a signal in a voice band of 300 Hz to 3.4 kHz and transmitted to the in-band ringer on the opposite side through a dedicated communication line. The opposite-side in-band ringer that has received this is configured to reproduce the original various signals or to function in the same manner as when receiving them.

[0003] When two-way communication is performed using a two-wire transmission line such as a balanced pair cable represented by an analog public telephone line, a device connected to the transmission line includes two lines.
It has a functional block for separating transmission / reception signals by using a hybrid transformer or the like serving as a line-to-line converter. In this case, since the user's own voice is delayed and returned through the two-wire to four-wire converter of the in-band ringer at the other end, the return component may become an echo and significantly impair the call. In order to prevent this, the in-band ringer is generally provided with an echo canceller, which estimates the characteristics of the echo path and will generate a 2-wire to 4-wire converter from the transmission signal. A pseudo echo is created by predicting the echo, and the echo is canceled by subtracting the pseudo echo from the received signal including the actual echo. In practice, after the transmission path is determined, training is performed, and the transmission characteristic of the echo path is corrected and optimized so that the residual echo obtained by subtracting the pseudo echo from the actual echo is minimized.

FIG. 4 shows a general communication path configuration example for realizing extension of an extension using an in-band ringer in a communication line that cannot transmit a DC signal.
A signal output from the terminal device 1 such as a telephone or a facsimile is transmitted to the exchange 4 via the terminal-side in-band ringer 2 and the station-side in-band ringer 3, and further transmitted to the station-side in-band ringer 5 and the terminal-side in-band ringer. 6 to the terminal device 7. Here, among the lines connecting the blocks, 8 to 11 are subscriber lines which can also transmit signals outside the voice band.
Between lines 5 and 6 and dedicated lines 12 and 13 whose band is limited.
The ringer signal, hooking information, and the like, which are signals outside the audio band, are transmitted through the dedicated line, and the signal in the audio band (in-band control signal) is composed of a combination of two different frequencies depending on the in-band ringer. ) And transmitted. In FIG. 4, the echo canceller is built in each of the four in-band ringers 2, 3, 5, and 6, and the echo components are canceled between 2 and 3 and between 5 and 6, respectively. I have.

FIG. 5 is a block diagram showing a configuration example of a conventional in-band ringer. In the figure, the station-side in-band ringer 5 and the terminal-side in-band ringer 6 are connected via a dedicated line 13. This dedicated line 13 is
This is a four-wire system with two lines and two down lines, and two lines are represented by one line. The station-side in-band ringer 5 is connected to the exchange 4 via a subscriber line 10, and the terminal-side in-band ringer 6 is connected to a terminal device 7 such as a telephone via a subscriber line 11. The subscriber lines 10, 11 are of a two-wire type, which is represented by a single line. The station-side in-band ringer 5 monitors a call current on the subscriber line 10, and controls a DC loop level and detects a polarity based on the call current, and a ringer signal (16H).
ringer signal detecting means 14 for detecting z), central control means 18 for controlling each part, and in-band control signal generating means 1 for generating a control signal by a combination of two different frequencies.
7, in-band signal detecting means 16 for detecting an in-band control signal transmitted via the dedicated line 13,
It comprises an echo canceller 19 that suppresses an echo generated in the 2-wire to 4-wire conversion part. On the other hand, the terminal-side in-band ringer 6 supplies a call current to the subscriber line 11,
DC generation means 21 for performing level detection and polarity control of the DC loop by the call current, central control means 24 connected to the DC generation means 21, in-band control signal generation means 23, and in-band control signal detection means 22 , A ringer signal generating means 20 for generating a ringer signal, and an echo canceller 25 for suppressing echo. It should be noted that the station-side in-band ringer 5 and the terminal-side in-band ringer 6 differ only in that they have the DC monitoring means 15 or the DC generation means 21. If they are configured to have both of them, the station The side in-band ringer 5 and the terminal-side in-band ringer 6 can have the same configuration. In the configuration described above, the polarity of the DC loop supplied from the exchange 4 is detected by the DC monitoring means 15, and the information is converted into an in-band control signal and transmitted to the terminal-side in-band ringer 6. On the other hand, in the terminal-side in-band ringer 6, the in-band control signal detecting means 22 detects the transmitted in-band control signal from the station-side in-band ringer 5 and reproduces a call current having the same polarity as the DC loop. . Further, the level of the DC loop that changes according to the on-hook / off-hook of the terminal device 7 is as follows:
The information is detected by the DC generator 21, converted into an in-band control signal, and transmitted to the station-side in-band ringer 5. On the other hand, in the station-side in-band ringer 5, the in-band control signal detecting means 16 detects the transmitted in-band control signal from the terminal-side in-band ringer 6 and detects the communication current so as to be at the same level as the DC loop. Control. The ringer signal supplied from the exchange 4 is
The information is detected by the ringer signal detecting means 14, the information is converted into an in-band control signal, and transmitted to the terminal-side in-band ringer 6. In the terminal side in-band ringer 6, the in-band control signal is detected by the in-band control signal detecting means 22, and a ringer signal is generated from the ringer signal generating means 20 at this time.

[0006] In such a transmission network, an echo path is formed in a 2-wire to 4-wire conversion section, so that a received signal from the 4-wire side becomes an echo and wraps around to a transmitted signal to the 4-wire side. Cause. Therefore, it is common to provide an echo canceller 25 to remove this echo. FIG.
FIG. 3 is a block diagram showing a configuration example of an echo canceller used for an in-band ringer. The echo canceller 25 includes a pseudo echo generation unit 27 and a synthesis unit 28. The two-wire to four-wire converter is shown as a hybrid circuit 26. Here the echo canceller 25
In general, a digital filter is used as the pseudo echo generation unit 27, and its filtering coefficient is set according to the transmission characteristics of the echo path. In practice, the transmission characteristics of the echo path change each time when a communication path is set for each call, so that the digital filter includes a digital signal processor (D
An adaptive filter whose filtering coefficient is variably controlled based on a control signal supplied from SP) is used, and the filtering coefficient is set to an optimum value every time a call is made. The echo canceller 25 generates a pseudo echo having a polarity opposite to that of the echo 29 wrapping around to the transmission side through the hybrid circuit 26 in the pseudo echo generating unit when the received signal is input, Supply to input. The combining unit 28 operates to add and cancel the pseudo echo to the transmission signal supplied from the other input terminal, thereby canceling out the echo 29 superimposed on the transmission signal. By the way, as described above, since the transmission characteristics of the echo path are different for each call, the echo canceller 25 receives a predetermined tone signal before the call, and uses the same tone signal as the echo 29 generated thereby to generate a pseudo echo generation unit. When the combining section 28 combines the pseudo echo generated at 27 with the pseudo echo, the operation of optimizing the filtering coefficient of the pseudo echo generating section 27 is performed so that the residual echo 30 output from the combining section 28 is minimized. . This is generally called training.

The relationship between the sequence from the no-call (standby) state to the end of the call and the timing of the echo canceller training will be described with reference to FIG. FIG. 7 shows the terminal-side in-band ringer 6 in FIG.
FIG. 4 is a DC loop voltage waveform diagram on the two-wire line accommodation side of FIG.
FIG. 9 is a diagram showing an example of a conventional incoming call sequence. t1 to t4 indicate the timing on the sequence. Before t1, a non-communication period, t1 to t2 are a calling period, t2 to t3 are a communication preparation period, t3 to t4 are a communication period, and after t4 is a non-communication period. . First, when there is no call, the in-band ringer supplies a +48 V DC signal to the telephone. Here, when there is a call from the opposite station, the inband ringer inverts the polarity and supplies a -48 V DC signal to the telephone as shown at timing t1 in the figure. When the polarity is inverted at t1, a ringing state is set, and a ringer signal is sent out during the period from t1 to t2, and the bell of the telephone rings. Then, when the receiver of the telephone is lifted (off-hook), a DC loop is formed and a communication current flows, so that the line voltage becomes low as shown at timing t2. When the terminal-side in-band ringer 6 detects this voltage change, the terminal-side in-band ringer 6 transmits the voltage change to the exchange 4 via the office-side in-band ringer 5, and the exchange 4
When this is confirmed, the polarity is inverted, and charging starts at the time of this polarity inversion. This polarity inversion information is transmitted to the terminal-side in-band ringer 6 via the station-side in-band ringer 5, and the terminal-side in-band ringer 6 detects the timing t3.
Then, the polarity is inverted as shown in FIG. In other words, since it is only about 100 to 300 mSec from the off-hook to the completion of the polarity reversal, the telephone can be talked before the telephone operator goes off-hook and starts talking. When the telephone call ends and the telephone receiver is lowered (on-hook), the DC loop is opened and the line voltage increases to + 48V. This voltage change is transmitted to the exchange 4 via the terminal-side in-band ringer 6 and the office-side in-band ringer 5, and the charging by the exchange is terminated and the state returns to the original non-communication state. In this series of sequences, the training of the conventional echo canceller was performed immediately after the polarity was inverted at t3 and the telephone became in a talk state, as shown in FIG. Since the echo to be canceled is an echo during a call, training must be performed after a call path has been established.
The DC loop from to t4 is being formed. The call preparation period from t2 to t3 differs depending on the type of the exchange, but is generally insufficient for training so short. For this reason, it was common to naturally train at the beginning of the call period.

[0008]

However, as described above, when training is performed at the beginning of a call period, the polarity inversion characteristics differ depending on the type of terminal device connected to the terminal-side in-band ringer. In the case of a terminal device that takes a long time to stabilize, training will be performed in the transition period of DC voltage fluctuation, and accurate echo estimation cannot be performed, and erroneous coefficients will be reflected in the adaptive filter, and echo There was a problem that cancellation was not performed properly. Also, for example, if training is performed after a time sufficient for the line voltage to stabilize,
It was highly likely that the call was started and the voice from the handset was mixed in the training signal and the training failed.

SUMMARY OF THE INVENTION The present invention has been made to solve such a problem, and accurately estimates an echo without being affected by a difference in polarity reversal characteristics depending on a type of a terminal device connected to a terminal-side in-band ringer. It is an object of the present invention to provide an in-speech signal device provided with a training means capable of performing the following.

[0010]

In order to solve the above-mentioned object, an in-voice signal apparatus according to the present invention comprises a training apparatus for optimizing the echo canceller in an in-voice signal apparatus having an echo canceller. It is performed during a call preparation period after the DC loop of the receiving terminal is formed and before the exchange switches the polarity.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail based on illustrated embodiments.

FIG. 1 is a block diagram showing one embodiment of an in-band ringer according to the present invention. In this figure, the point different from the conventional one is that a central control means 31 is provided with a timer function (not shown), and the other configuration is the same as that of FIG. 3 showing the conventional configuration. The timer starts its operation at the same time that the DC monitoring means 21 detects that the terminal device 7 has gone off-hook, and is a preset time, specifically, a time required for training the echo canceller, and counts, for example, 500 mSec. Things.

The operation of the in-band ringer according to the present invention will be described below. FIG. 2 is a DC loop voltage waveform diagram on the two-wire line accommodating side of the terminal-side in-band ringer 6 in FIG. 1, and is a diagram showing the relationship between the incoming call sequence and the training timing of the echo canceller. In the figure, first, the non-communication state of the line voltage +48 V is continued up to the point indicated by t1. The terminal-side in-band ringer 6 detects a call from the opposite station, reverses the polarity at the timing of t1, and enters a call state. Then, when the called terminal device 7 is taken off-hook, a DC loop is formed and a call current flows, and the timing t2
As shown in FIG. When the terminal-side in-band ringer 6 detects the off-hook of the terminal device 7, the timer in the central control means 31 starts counting a predetermined time and transmits the off-hook information to the upstream exchange 4. After detecting the off-hook, the echo canceller starts training. When the exchange 4 detects off-hook information, that is, a DC loop from the local in-band ringer 5 in the subscriber line 10, it stops calling and inverts the polarity of the DC loop. The office-side in-band ringer 5 receives the polarity inversion of the DC loop from the exchange 4, and transmits the polarity inversion information to the terminal-side in-band ringer 6 as an in-band signal. When the terminal-side in-band ringer 6 receives the polarity inversion signal sent via the dedicated line 13 as an in-band signal, it checks the state of the timer in the central control means 31. If the timer has counted out for a predetermined time, for example, 500 ms, and has timed out, the polarity of the subscriber line 11 accommodating the terminal device 7 is immediately inverted as shown at t3. Further, if the timer is counting, after the timeout, the polarity of the subscriber line 11 accommodating the terminal device 7 is inverted as shown at t3. When the polarity inversion is performed, a talking period and a talking or data transmission are performed, and the operation is continued until the on-hook indicated at t4 is performed. When on-hook is performed at t4, the DC loop of the subscriber line 11 is released, and the line voltage becomes +48 V, which is a non-communication period.

Next, a series of procedures shown in FIG. 2 will be described in more detail with reference to a flowchart. FIG. 3 is a diagram showing a flow chart of the incoming call system of the terminal-side in-band ringer according to the present invention, which will be described in order for each STEP number attached to the drawing. First, the terminal device accommodated by the terminal-side in-band ringer is in an on-hook state, and the terminal-side in-band ringer is in a standby state in which a call or a call is monitored. (STEP 1) When an in-band signal, which is call information from the opposing station-side in-band ringer, is detected, the operation proceeds to the next STEP 3.
(STEP 2) Upon detecting the call, the terminal-side in-band ringer reverses the polarity of the subscriber line accommodating the terminal device and calls the terminal device by transmitting a ringer signal or the like. (STEP 3) If the called terminal device goes off-hook and responds, the next S
Move on to TEP5. (STEP 4) The terminal-side in-band ringer that has detected the off-hook of the terminal device starts counting by a timer. (STEP 5) In order to transmit the off-hook information of the terminal device to the exchange, an in-band signal indicating off-hook is transmitted to the station-side in-band ringer. (STEP 6) After detecting the off-hook of the terminal device, the training of the echo canceller is started. (STEP 7) When it is detected from the in-band signal from the station-side in-band ringer that the exchange has reversed the polarity and started charging, the next S is executed.
Move to TEP8. (STEP 8) If the timer has finished counting the predetermined time and has timed out when the accounting information is detected, the next STEP
Move to 10, if the timer is still counting, wait for timeout, and after timeout, the next STEP
Move to 10. (STEP 9) The terminal-side in-band ringer reverses the polarity of the subscriber line when both the receipt of the billing information from the exchange and the completion of the counting by the timer are satisfied, and bills the terminal device. Communicate information. (STEP 10) After inverting the polarity of the subscriber line, the communication state is established, and the terminal-side in-band ringer relays the bidirectional transmission signal.
(STEP 11) When the terminal device on the calling side or the called side goes on-hook, the line is disconnected and the call is terminated, and the process proceeds to the next STEP 13. (STEP 12) When the line is disconnected and the call ends, the process returns to the standby state similar to STEP 1. (STEP13)

As described above, the characteristic feature of the present invention is that the training time is secured during the call preparation period,
To provide training during the call preparation period. That is,
The conventional terminal-side in-band ringer transmits the sequence received from the exchange as it is.In the present invention, if the call preparation period is shorter than a predetermined time in the sequence received from the exchange, the communication preparation period is extended and transmitted. In order to carry out the training of the echo canceller, sufficient time is secured.

As described above, since the in-voice signal apparatus according to the present embodiment performs the echo canceller training during the call preparation period, the polarity inversion characteristic varies depending on the type of terminal connected to the terminal-side in-band ringer. It is possible to accurately estimate the echo without being affected by the difference,
Training can be performed without waiting until after the line voltage is stabilized.

[0017]

According to the present invention, even if the polarity reversal characteristics are different depending on the type of the terminal device connected to the terminal side in-band ringer, the training is performed before the polarity reversal, so that it is not affected, and accurate echo estimation can be performed. This makes it possible to properly cancel the echo. In addition, the training is performed at an earlier timing than in the related art, so that the probability of training failure due to the mixture of speech signals and the like is reduced.

[Brief description of the drawings]

FIG. 1 is a block configuration example of an in-band ringer on a station side and a terminal side based on the present invention.

FIG. 2 is a timing diagram showing echo canceller training of a terminal-side in-band ringer on the called side in the configuration of an in-voice signal device for transmitting charging information (polarity inversion signal) according to the present invention.

FIG. 3 is a flowchart example of the incoming call system of the terminal-side in-band ringer according to the present invention.

FIG. 4 is a configuration example in the case where an in-voice signal device is used by extending an extension.

FIG. 5 is a block configuration example of a conventional in-band ringer on the station side and on the terminal side.

FIG. 6 is a configuration example of an echo canceller used for an in-band ringer.

FIG. 7 is a timing diagram showing echo canceller training of a terminal-side in-band ringer on the called side in a conventional in-voice signal device configuration for transmitting billing information (polarity inversion signal).

[Description of sign]

1, 7: terminal device 2, 6: terminal-side in-band ringer (signal device in voice band) 3, 5: station-side in-band ringer (signal device in voice band) 4: switch 8 11 to subscriber line 12, 13 to dedicated line 14 to ringer signal detecting means 15 to DC monitoring means 16, 22 to in-band signal detecting means 17, 23 to in-band Signal generation means 18, 24 Central control means 19, 25 Echo canceller 20 Ringer signal generation means 21 DC generation means 26 Hybrid circuit (2-wire to 4-wire conversion unit) 27 ... Pseudo echo generator 28 ... Synthesizer 29 ... Echo 30 ... Residual echo 31 ... Central control means

Continuation of the front page (72) Inventor Atsushi Fujisawa 2-1-1 Kotani, Samukawa-cho, Koza-gun, Kanagawa F-term in Toyo Tsushinki Co., Ltd. (Reference) 5K046 BB01 EF02 GG00 HH05 HH11 HH79 PP06 5K050 AA01 BB06 BB11 CC03 DD21 DD26 DD30 EE04 FF12 FF17 GG04 GG13 HH10

Claims (1)

[Claims] In an in-voice signal device provided with an echo canceller, training for optimizing the echo canceller is performed during a call preparation period after a DC loop is formed at a terminating terminal until the polarity of the exchange is inverted. A signal device in a voice band, characterized in that: