JPH02230824A - Echo suppressor - Google Patents

Abstract

PURPOSE:To guarantee the adaptive operation of an adaptive filter by giving an error signal being the result of multiplying a prescribed weight with a prescribed constant to the adaptive filter when an output signal of a subtractor is larger than an output signal from a residual echo detection means and the result of multiplying a prescribed weight with a constant zero to the adaptive filter when smaller. CONSTITUTION:When an absolute value of an output signal level from a subtractor 152 is larger than an absolute value of an output signal level from a computing element 172, a weighted constant 2a given to a weighting circuit 19 is selected by a constant selection circuit 212, it is regarded that the reception signal is cancelled by the computing element 172. When the absolute value of an output signal level from a subtractor 152 is smaller than the absolute value of an output signal level from the computing element 172, a weighted constant 0 given to the weighting circuit 19 is selected by the constant selection circuit 212, it is regarded that the reception signal is cancelled by the computing element 172. Thus, the adaptive operation of the adaptive filter 20 is warranted.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an echo canceling device used in realizing two-wire bidirectional digital transmission, and in particular, to ensuring adaptive operation of the adaptive filter used. The present invention relates to an echo canceling device.

[Conventional technology]

Conventionally, this type of echo canceling device uses a pseudo-echo signal generated by the adaptive operation of an adaptive filter that employs a sine algorithm on the 4-wire side of a 2-wire 4-wire conversion circuit to remove signals from a transmitting circuit to a receiving circuit. It is known as a device that removes echo signals leaking into the air.

FIG. 2 is a circuit diagram showing such a conventional echo canceling device.

A transmitter 1 provided with an input terminal Ti is connected to a two-wire transmission line 3 via a hybrid transformer 2. This hybrid transformer 2 is connected to a receiving section 6 via a low-pass filter 4 and an echo removing means 5. The receiving section 6 is provided with an output terminal TO.

The echo removing means 5 removes the pseudo echo signal S. A digital/analog (DA) converter 5 that takes in and converts it into an analog signal.
1 and analog pseudo echo signal A from the DA converter 51
A subtracter 5 that removes the echo signal by subtracting the SDE and the mixed signal St in which echo compensation and received signals leak from the transmitter 1 to the receiver 2 on the 4-wire side of the 2-wire 4-wire converter.
It is composed of 2. The residual echo detection means 7 includes a delay circuit 71 that delays the output signal from the subtracter 52 by a predetermined time, and a difference or sum of the delayed signal from the delay circuit 71 and the output signal from the echo removal means 50 and the subtracter 52. It is configured to include a computing unit 72 that performs calculations. The polarity detector 8 is adapted to determine the polarity of the output signal from the arithmetic unit 72 of the residual echo detection means 7. The output signal from the polarity detector 8 is taken into a weighting circuit 9, and the weighting circuit 9 obtains an error signal by multiplying it by a constant. The adaptive filter 10 is configured so that its tap coefficients can be switched using the error signal from the weighting circuit 9, and is adapted to take in transmission data and adaptively generate a pseudo echo signal. As such an echo canceling device, for example, the Japanese Patent Publication No. 61-7
There is No. 3429.
HOrlV. According to the echo canceling device as described above, the mixed signal Sl, which is a mixture of echo signals and received signals leaking from the transmitter 1 to the receiver 2 on the 4-wire side of the 2-wire 4-wire converter, is processed by the subtracter 52. , and the analog pseudo echo signal ASDE from the DA converter 51. The output signal from the subtracter 52 is given to the receiving section 6 and the residual echo detecting means 7.

In the residual echo detection means 7 , the output signal from the subtracter 52 is delayed by a predetermined time by a delay circuit 71 and then supplied to an arithmetic unit 72 . Furthermore, in the arithmetic unit 72, the delay circuit 7
1 and the output signal from the subtracter 52 of the echo canceling means 50. The output signal from the arithmetic unit 72 is subjected to polarity detection by a polarity detector 8 and is applied to a weighting circuit 9 . In the weighting circuit 9, the polarity detector 8
The output signal from is multiplied by a constant 2α and output as an error signal. The error signal output from the weighting circuit 9 switches the tap coefficients of the adaptive filter 10.

By the way, when the adaptive filter 10 is operated adaptively as described above, it is necessary to accurately detect the polarity of the residual echo signal (the echo signal plus the pseudo echo signal) by the polarity detector 8. However, the subtractor 52
The output signal of , of course, includes the received signal, so if the output signal of the subtractor 52 is directly input to the scratchiness detector 8, the residual echo signal level will be about the same as the received signal level. The output signal from the polarity detector 8 no longer accurately determines the polarity of the residual echo signal. As a result, the adaptive filter 10 loses its adaptive ability.

Therefore, in the conventional echo canceling device, attention is paid to the characteristics of the eye pattern of the received signal, and a residual echo detection means 7 is provided to prevent the probability that the received signal will be canceled to zero. That is, according to the characteristics of the receiving eye pattern of a transmission line code including a binary code system, the current sample value and the sample value n samples (n is a positive integer) before are almost the same value or have opposite polarity. The minimum values of the probabilities that the absolute values are almost the same will be canceled by the probability of a certain positive value that is not zero. Therefore, by detecting the polarity of the sum or difference between the current sample value, which is the output signal from the subtracter 52, and the sample value n samples before (n is a positive integer) obtained via the delay circuit 71, Since the sign of the residual echo signal can be detected with a probability of a certain positive value other than zero, the adaptive operation of the adaptive filter 10 is guaranteed.

[Problem to be solved by the invention]

However, the above-mentioned conventional echo canceling device cannot detect the polarity of the residual echo signal even if the received signal is not canceled, even if the output signal of the polarity detector 8 does not necessarily match the polarity of the residual echo signal. The coefficients of the adaptive filter 10 are sequentially updated according to the following. Therefore, if the positive and negative probabilities of the polarity detector 8 when the received signal is not canceled are not equal, the adaptive
The disadvantage is that the adaptive operation of the filter 10 is no longer guaranteed.

The present invention has been made in order to eliminate the above-mentioned drawbacks, and an object of the present invention is to provide an echo cancellation device that guarantees adaptive operation of an adaptive filter.

[Means to solve the problem]

The echo canceling device of the present invention detects the echo signal by calculating the difference between the mixed signal, which is a mixture of the echo signal leaking from the transmitting circuit to the receiving circuit and the received signal, and the pseudo echo signal on the 4-wire side of the 2-wire 4-wire converter. a residual echo detecting means for delaying the output signal from the echo removing means by a predetermined time and calculating the difference or sum between the delayed signal and the output signal from the echo removing means; A polarity detector that determines the polarity of the output signal from the echo detection means, a weighting circuit that obtains an error signal by multiplying the output signal from this polarity detector by a constant, and a filter tap using the error signal from this weighting circuit. An echo cancellation device comprising an adaptive filter that adaptively generates a pseudo echo signal based on transmitted data by switching coefficients,
The weighting circuit compares the levels of the output signal from the residual echo detection means and the output signal from the echo removal means, and when the level of the output signal from the echo removal means is greater than the level of the output signal from the residual echo detection means. The invention is characterized in that it is provided with a control means for switching the constant to zero.

The present invention includes a second filter that has a plurality of coefficients and whose output can be adjusted, and a tap switching means that can adjust each coefficient and output according to the shift in transmission/reception timing, thereby making it possible to remove echo signals. can.

〔Example〕

Next, the present invention will be explained with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of an echo canceling device of the present invention. In FIG. 1, the echo canceling device includes a transmitter 11, a hybrid transformer 12, a two-wire transmission line 13, a low-pass filter 14, an echo canceller 15, a receiver 16, and a residual echo detector. 17, a polarity detector 18, a weighting circuit 19, an adaptive filter 20, and a control means 21.

A transmitter 11 provided with an input terminal Ti is connected to a two-wire transmission line 13 via a hybrid transformer 12.

The hybrid transformer 12 has a low-pass filter 1
4. Connected to the receiving section 16 via the echo canceling means 1!1. The receiving section 16 is provided with an output terminal To.

The echo removing means 15 generates pseudo echo signals SI,! A digital/analog (DA) converter 151 takes in AS. as an analog signal, and an analog pseudo echo signal AS. and transmitter 1 on the 4-wire side of the 2-wire 4-wire converter.
The subtracter 152 subtracts a mixed signal S, in which an echo signal leaking into the receiving section 16 and a received signal is mixed, from 1 to remove the echo signal. The residual echo detection means 17 includes a delay circuit 171 that delays the output signal from the subtracter 152 by a predetermined time, and a difference or sum between the delayed signal from the delay circuit 171 and the output signal from the subtracter 152 of the echo removal means 15. The computer is configured to include an arithmetic unit 172 that performs calculations. The polarity detector IB is configured to determine the polarity of the output signal from the arithmetic unit 172 of the residual echo detection means 17. The weighting circuit 19 obtains an error signal by multiplying the output signal from the polarity detector 18 by a constant signal. The adaptive filter 20 is configured so that tap coefficients of the filter can be switched using an error signal from the weighting circuit 19, and is adapted to take in transmission data and adaptively generate a pseudo echo signal. The control means 21 compares the level of the output signal from the subtracter 152 and the output signal from the residual echo detection means 7, and the absolute value of the output signal level from the subtractor 152 is the output signal of the residual echo detection means 17. It consists of a level comparator 211 that outputs a switching signal when the level is greater than the absolute value, and a constant selection circuit 212 that switches the constant to zero based on the switching signal output from the level comparator 211.

The operation of the embodiment configured in this way will be explained.

The binary data series input to the input terminal Ti is transmitted to the transmitter 1
1 and an adaptive filter 20. The binary data sequence is converted into a transmission line code in the transmitter 11, and then sent to the two-wire transmission line l3 via the hybrid 1 transformer l2.

Here, due to the impedance mismatch of the two-wire transmission line 13, the output signal of the transmitting fls11 leaks into the receiving circuit side as an echo signal and is supplied to the low-pass filter 14.

On the other hand, the received signal is also supplied from the two-wire transmission line l3 to the low-pass filter 14 via the hybrid transformer 12. The mixed signal (the echo signal plus the received signal) from which unnecessary high frequency components have been suppressed by the low-pass filter 14 is supplied to the subtracter l52. Here, the pseudo echo signal Sllll generated by the adaptive filter 20
l! is converted into a pseudo echo signal ASox, which is an analog signal, by the DA converter 151 and input to the subtracter 152 . Therefore, if the difference signal (mixed signal - pseudo echo signal ASD!) which is the output of the subtracter 152 is sufficiently small compared to the received signal, it will be accurately demodulated in the receiver 16 and received from the output terminal To. A value data series appears.

Here, a closed loop consisting of the DA converter 151, the subtracter 152, the arithmetic unit 172 of the residual echo detection means 7, the polarity detector 18, the weighting circuit 19, and the adaptive filter 20 realizes the adaptive operation of the adaptive filter 20. It is something.

Here, the absolute value of the output signal level of the subtracter 152 and the absolute value of the output signal level of the arithmetic unit 172 of the residual echo detection means 17 are compared by the level comparator 211.
When the output signal level from the subtracter 152 is higher than the output signal level from the subtracter 152, the constant 2α is
When the output signal level from the subtracter 152 is smaller than the output signal level from the subtracter 152, a switching signal for selecting the constant zero is output. The constant selection circuit 212 inputs the constant 2α or the constant zero as an error signal to the adaptive filter 20 in response to the switching signal. Thereby, the adaptive filter 20 performs tap switching based on the error signal and performs an adaptive operation to generate the pseudo echo signal S. ffi will be generated.

In this way, in this embodiment, the level comparator 211 compares the output signal level from the subtracter 152 and the output signal level from the arithmetic unit 172, and the absolute value of the output signal level from the subtracter 152 is determined by the arithmetic unit 172. If the absolute value of the output signal level is greater than the absolute value of the output signal level, the received signal is canceled by the arithmetic unit 172, and the constant selection circuit 212 selects 2α as the weighting constant to be given to the weighting circuit 19. Also,
In this embodiment, the level comparator 211 compares the output signal level from the subtracter 152 and the output signal level from the arithmetic unit 172, and the absolute value of the output signal level from the subtracter 152 is determined by the output signal level from the arithmetic unit 172. If it is smaller than the absolute value of the signal level, the arithmetic unit 172 assumes that the received signal is not canceled and selects a weighting constant of zero to be given to the weighting circuit 19 in the constant selection circuit 1a2 1 2, and
Give to 9. By doing so, adaptive operation of the adaptive filter 20 can be guaranteed.

〔Effect of the invention〕

As described above, the present invention compares the level of the output signal of the subtracter and the output signal from the residual echo detection means, and when the output signal of the subtractor is larger than the output signal from the residual echo detection means, a predetermined constant is set. When the error signal is adaptively filtered with a constant zero when it is small, the error signal is adaptively filtered.
By applying this to the filter, there is an effect that the adaptive operation of the adaptive filter can be guaranteed.

[Brief explanation of the drawing]

FIG. 1 is a circuit diagram showing an embodiment of the present invention, and FIG. 2 is a circuit diagram showing a conventional device. 11...Transmission unit, 2...Hybrid transformer, 3...
・2-wire transmission line, 5...Echo removal means, 16...Receiving section, 7...Residual echo detection means, 8...Polarity detection device, 19... weighting circuit, 0... adaptive filter, 1...
...control means.

Claims (1)

[Claims] an echo removing means that removes the echo signal by calculating the difference between the mixed signal containing the echo signal leaking from the transmitting circuit to the receiving circuit on the 4-wire side of the 2-wire 4-wire converter and the pseudo echo signal; residual echo detection means for delaying the output signal from the echo removal means by a predetermined time and calculating the difference or sum between the delayed signal and the output signal from the echo removal means; and the output signal from the residual echo detection means. a polarity detector that determines the polarity of the polarity detector, a weighting circuit that obtains an error signal by multiplying the output signal from this polarity detector by a constant, and a filter tap coefficient that is switched by the error signal from this weighting circuit to transmit data. and an adaptive filter that adaptively generates a pseudo echo signal based on the residual echo detection means. An echo removal device comprising: control means for switching a constant of the weighting circuit to zero when the level of the output signal from the residual echo detection means is higher than the level of the output signal from the residual echo detection means.