CN112037808B - Echo elimination method and device for elevator car - Google Patents

Abstract

The present invention relates to an echo cancellation method for an elevator car, comprising the following steps: S1: a far-end audio signal is input to an MCU through an audio input interface, and then a current echo signal is generated by a loudspeaker after processing; S2: the far-end audio signal is also input to an adaptive filter, and a current echo estimation signal is output; S3: a microphone collects a near-end audio signal, a current echo signal, and a previous round of echo signals; S4: according to an echo cancellation algorithm pre-set in the MCU, the signal collected by the microphone, the current echo estimation signal, and the previous round of echo estimation signals are calculated to obtain a pure echo signal, which is output by an audio output interface; S5: the current echo estimation signal is saved and used as a reference parameter for the next round of echo cancellation signals. The technical solution of the present invention adopts a structure in which a loudspeaker and a microphone are placed vertically, which can reduce the generation of echoes, and at the same time, the loudspeaker faces the floor of the car, and the distance between the loudspeaker and the floor is far, which can reduce the generation of echoes.

Description

Echo elimination method and device for elevator car

Technical Field

The invention belongs to the technical field of echo elimination, and in particular relates to an echo elimination method and device for an elevator car.

Background technique

In the field of elevator IoT, when someone is trapped in an elevator, rescuers will use voice to comfort and communicate with the trapped person. However, since the elevator car is a closed space with a small space, a large echo will be caused during the two-way voice process, resulting in poor voice effect. There are great obstacles for rescuers to communicate with people inside the car, making rescue difficult.

Conventional acoustic echo cancellation (AEC) is used to eliminate unwanted echoes in hands-free calls. An adaptive filter is usually used to model the echo path pulses and remove the echo from the microphone output signal.

However, due to the special environment of the elevator car, there are multiple echoes of the current echo and the previous echo. The intensity of the echo when it is received by the microphone is different from the sound played by the speaker. Therefore, traditional echo cancellation technology will have inaccurate calculations, resulting in the inability to accurately remove the echo signal and delay the implementation of the elevator rescue operation.

Summary of the invention

In order to solve the above problems, the present invention provides an echo elimination method and device for an elevator car. Through the special placement of the microphone and the speaker in the device and the internal inheritance of the independently developed dual echo elimination algorithm, the echo can be eliminated and the call quality can be improved. It is well used in the elevator Internet of Things.

An echo cancellation method for an elevator car comprises the following steps:

S1: The remote audio signal is input to the MCU through the audio input interface and modulated by the speaker to generate the current echo signal;

S2: The far-end audio signal is also input into the adaptive filter, and the adaptive filter calculates and outputs a current echo estimation signal;

S3: The microphone collects the near-end audio signal, the current echo signal generated by the speaker in step S1, and the previous round of echo signal generated by the speaker in the previous round of echo cancellation;

S4: According to the echo cancellation algorithm pre-set in the MCU, the signal collected by the microphone in step S3, the current echo estimation signal in step S2, and the previous round echo estimation signal stored in the MCU are calculated to obtain a pure echo signal, which is output through the audio output interface;

S5: Save the current echo estimation signal for use as a reference parameter for the next round of echo cancellation signal.

Preferably, the calculation formula of the current echo estimation signal in step S2 is:

Where _r'n is the current echo estimation signal, ^wTn _is the current coefficient matrix, and _yn is the far-end input signal.

Preferably, the process of obtaining w ^T _n is:

Calculate the echo estimation error using the following formula: _en = _dn – (λr'n _-1 + _r'n ), where _dn = _xn + _rn + rn _-1 ;

Through iterative calculation, minJ = e _n ² is obtained, where the iteration step is: The iteration equation is W _n+1 = w _n + η _n * _en * _yn ;

When J is the minimum value, the current coefficient matrix w ^T _n is obtained through the above iterative calculation.

Among them, _en is the echo estimation error, _dn is the sum of the audio signals collected by the microphone in step S3, r'n _-1 is the previous round of echo estimation signal, _r'n is the current echo estimation signal, _xn is the near-end input signal, _rn is the current echo signal, rn _-1 is the previous round of echo signal, J is a reference symbol with a value of e _n ² , _ηn is the iteration step size, ^yTn _is the transposed matrix of the far-end input signal _yn , Wn ₊₁ is a reference symbol that stores the value of _wn + _ηn * _en * _yn , and λ is the weakening coefficient.

Preferably, the calculation formula of the pure echo signal in step S4 is: _un =( _xn + _rn +rn _-1 )-(λr'n _-1 + _r'n ), wherein _un is the pure echo signal, r'n _-1 is the previous round echo estimation signal, _r'n is the current echo estimation signal, λ is the weakening coefficient, _xn is the near-end input signal, _rn is the current echo signal, and rn _-1 is the previous round echo signal.

Preferably, the preservation of the current echo estimation signal in step S5 is expressed in the algorithm as setting r _n =r _n-1 and using the current echo estimation signal as a weakening parameter of the previous round of echo in the next round.

The present invention also provides an echo cancellation device for an elevator car, comprising:

Audio input interface: used to receive remote audio signals;

Speaker: used to generate echo signal;

Adaptive filter: used to generate an echo estimation signal based on the far-end audio signal;

Microphone: used to receive echo and near-end audio signals;

MCU: used to store the echo cancellation algorithm, receive and process the audio signals output by the above devices, and obtain pure echo;

Audio output interface: used to output pure echo signal.

Preferably, the speaker is perpendicular to the floor of the car, and the speaker microphone is placed vertically.

Preferably, it also includes an audio encoder and an audio decoder. The audio encoder decodes the received audio signal into a digital signal and sends it to the MCU for processing. The audio signal processed by the MCU is decoded by the audio decoder and then output.

The beneficial effects of the present invention are as follows: the technical solution of the present invention adopts a structure in which the loudspeaker and the microphone are placed vertically, which can reduce the generation of echoes. At the same time, the loudspeaker faces the floor of the car, and the distance between the loudspeaker and the ground is relatively far, which can weaken the generation of echoes.

The present invention adopts double echo cancellation in the algorithm. Due to the special environment of the car, there are multiple echoes of the current echo and the previous echo mixed in the echo. The algorithm used in the present invention adopts double cancellation to minimize the echo. At the same time, the algorithm considers that the echo is different in intensity when it is received by the microphone and the sound played by the speaker. The echo weakening coefficient is added when estimating the echo path to better fit the echo and minimize the final echo.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flow chart of the method provided by the present invention.

FIG. 2 is an internal connection block diagram of the device provided by the present invention.

FIG. 3 is an external structural diagram of the device provided by the present invention.

Among them, 1. Audio input interface; 2. Speaker; 3. Microphone; 4. Audio output interface.

Detailed ways

The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

As shown in FIG1 , an embodiment of the present invention provides a method for eliminating echo in an elevator car, comprising the following steps:

1. The remote audio signal is input to the MCU through the audio input interface and modulated, and then the speaker generates the current echo signal.

2. The far-end audio signal is also input to the adaptive filter, which calculates and outputs the current echo estimation signal. The calculation formula of the current echo estimation signal is:

Where _r'n is the current echo estimation signal, ^wTn _is the current coefficient matrix, _yn is the far-end input signal, _wn and _yn represent the cumulative sum, wn is n _and takes the value from 0 to N-1, and _{the same is} true for yni.

The process of obtaining w ^T _n is:

Calculate the echo estimation error using the following formula: _en = _dn – (λr'n _-1 + _r'n ), where _dn = _xn + _rn + rn _-1 ;

Through iterative calculation, minJ = e _n ² is obtained, where the iteration step is: The iteration equation is:

Wn ₊₁ = _wn + _ηn * _en * _yn ;

When J is the minimum value, the current coefficient matrix w ^T _n is obtained through the above iterative calculation.

Among them, _en is the echo estimation error, _dn is the sum of the audio signals collected by the microphone in step S3, r'n _-1 is the previous round of echo estimation signal, _r'n is the current echo estimation signal, _xn is the near-end input signal, _rn is the current echo signal, rn _-1 is the previous round of echo signal, J is a reference symbol with a value of e _n ² , _ηn is the iteration step size, ^yTn _is the transposed matrix of the far-end input signal _yn , Wn ₊₁ is a reference symbol that stores the value of _wn + _ηn * _en * _yn , and λ is the weakening coefficient.

3. The microphone collects the near-end audio signal, the current echo signal generated by the speaker in step S1, and the previous round of echo signal generated by the speaker in the previous round of echo cancellation.

4. According to the echo cancellation algorithm, the signal collected by the microphone in step S3, the current echo estimation signal in step S2, and the previous round echo estimation signal stored in the MCU are calculated to obtain a pure echo signal, which is output by the audio output interface. The calculation formula of the pure echo signal is: _un = ( _xn + _rn + _rn-1 ) - (λr'n _-1 + _r'n ), where _un is the pure echo signal, r'n _-1 is the previous round echo estimation signal, _r'n is the current echo estimation signal, λ is the weakening coefficient, _xn is the near-end input signal, _rn is the current echo signal, and rn _-1 is the previous round echo signal.

5. Save the current echo estimation signal as a reference parameter for the next round of echo cancellation signal. The current echo estimation signal is saved in the algorithm by setting r _n = r _n-1 and using the current echo estimation signal as a weakening parameter of the previous round of echo in the next round.

As shown in Figures 2 and 3, an embodiment of the present invention also provides an echo cancellation device for an elevator car, including an audio input interface 1 for receiving a far-end audio signal, a speaker 2 for generating an echo signal, an adaptive filter for generating an echo estimation signal according to the far-end audio signal, a microphone 3 for receiving an echo and a near-end audio signal, an MCU for storing an echo cancellation algorithm, receiving various audio signals output by the above devices for processing to obtain a pure echo, and an audio output interface 4 for outputting a pure echo signal.

Since the input audio is not data that the MCU can directly process, the device also includes an audio encoder and an audio decoder. The audio encoder decodes the received audio signal into a digital signal and sends it to the MCU for processing. The audio signal processed by the MCU is decoded by the audio decoder and output.

As an embodiment of the present invention, the speaker is perpendicular to the floor of the car, and the speaker microphone is placed vertically.

The working principle of the device provided in the embodiment of the present invention is shown in the above-mentioned method for echo elimination of an elevator car.

Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can still modify the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. They should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (5)

1. A method for eliminating echo in an elevator car, comprising the following steps: S1: The remote audio signal is input to the MCU through the audio input interface and then processed by the speaker to generate the current echo signal; S2: The far-end audio signal is also input into the adaptive filter, and the adaptive filter calculates and outputs a current echo estimation signal; S3: The microphone collects the near-end audio signal, the current echo signal generated by the speaker in step S1, and the previous round of echo signal generated by the speaker in the previous round of echo cancellation; S4: According to the echo cancellation algorithm pre-set in the MCU, the signal collected by the microphone in step S3, the current echo estimation signal in step S2, and the previous round echo estimation signal stored in the MCU are calculated to obtain a pure echo signal, which is output through the audio output interface; S5: Save the current echo estimation signal as a reference parameter for the next round of echo cancellation signal; The calculation formula of the current echo estimation signal in step S2 is: Where _r'n is the current echo estimation signal, ^wTn _is the current coefficient matrix, _yn is the far-end input signal, _wn and _yn represent the cumulative sum, wn is n _and takes the value from 0 to N-1, and _{yni is} the same; The acquisition process of w ^T _n is: The echo estimation error is calculated using the formula: _en = _dn - (λr'n _-1 + _r'n ), where _dn = _xn + _rn + rn _-1 ^; min J = _en2 is obtained through iterative calculation, where the iteration step is: The iteration equation is W _n+1 = w _n + η _n * _en * _yn ; When J is the minimum value, the current coefficient matrix w ^T _n is obtained through the above iterative calculation; Among them, _en is the echo estimation error, _dn is the sum of the audio signals collected by the microphone in step S3, r'n _-1 is the previous round of echo estimation signal, _r'n is the current echo estimation signal, _xn is the near-end input signal, _rn is the current echo signal, rn _-1 is the previous round of echo signal, J is a reference symbol with a value of e _n ² , _ηn is the iteration step size, ^yTn _is the transposed matrix of the far-end input signal _yn , Wn ₊₁ is a reference symbol that stores the value of _wn + _ηn * _en * _yn , and λ is the weakening coefficient. 2. The echo cancellation method for an elevator car according to claim 1, characterized in that the calculation formula of the pure echo signal in step S4 is: _un =( _xn + _rn +rn _-1 )-(λr'n _-1 + _r'n ), wherein _un is the pure echo signal, r'n _-1 is the previous round echo estimation signal, _r'n is the current echo estimation signal, λ is the weakening coefficient, _xn is the near-end input signal, _rn is the current echo signal, and rn _-1 is the previous round echo signal. 3. The method for cancelling the echo of an elevator car according to claim 2, characterized in that the preservation of the current echo estimation signal in step S5 is expressed in the algorithm as setting r _n = r _n-1 and using the current echo estimation signal as a weakening parameter of the previous round of echo in the next round. 4. An echo cancellation device for an elevator car, using the echo cancellation method according to claim 1, characterized in that it comprises: Audio input interface: used to receive remote audio signals; Speaker: used to generate echo signal; Adaptive filter: used to generate an echo estimation signal based on the far-end audio signal; Microphone: used to receive echo and near-end audio signals; MCU: used to store the echo cancellation algorithm, receive and process the audio signals output by the audio input interface, speaker, adaptive filter and microphone to obtain a pure echo; Audio output interface: used to output pure echo signal; The loudspeaker is perpendicular to the floor of the car, and the loudspeaker microphone is placed vertically. 5. According to the echo cancellation device for an elevator car as described in claim 4, it is characterized in that it also includes an audio encoder and an audio decoder, the audio encoder decodes the received audio signal into a digital signal and sends it to the MCU for processing, and the audio signal processed by the MCU is decoded by the audio decoder and output.