CN113409755B

CN113409755B - Active noise reduction method and device and active noise reduction earphone

Info

Publication number: CN113409755B
Application number: CN202110846482.2A
Authority: CN
Inventors: 徐银海; 刘益帆
Original assignee: Beijing Ansheng Haolang Technology Co ltd
Current assignee: Beijing Ansheng Haolang Technology Co ltd
Priority date: 2021-07-26
Filing date: 2021-07-26
Publication date: 2023-10-31
Anticipated expiration: 2041-07-26
Also published as: CN113409755A

Abstract

The application provides an active noise reduction method and device and an active noise reduction earphone. The method comprises the following steps: collecting an original noise signal by a reference microphone; collecting an error signal through an error microphone, wherein the error signal comprises an error between an original noise signal transmitted to a spatial point where the error microphone is located and a noise reduction signal, and the noise reduction signal is determined according to the original noise signal and an initial filter coefficient; inputting the error signal into a non-causal secondary path inverse estimation unit to obtain a pre-estimated error signal, wherein the secondary path inverse estimation unit is used for simulating the inverse of a secondary path; and determining the filter coefficient according to the original noise signal and the estimated error signal. According to the active noise reduction method provided by the application, the error signal is actively advanced in the process of adaptively adjusting the filter coefficient so as to accurately correspond to the original noise signal in time, so that the adverse effect caused by the secondary path can be overcome, and the stability of the active noise reduction process is improved.

Description

Active noise reduction method and device and active noise reduction earphone

Technical Field

The application relates to the technical field of acoustics, in particular to an active noise reduction method and device and an active noise reduction earphone.

Background

In recent years, with a great increase in market demand, active noise reduction (ANC, active Noise Cancellation) headphones are gaining more and more attention. The design core of the active noise reduction earphone is a filter, such as a feedforward filter W _ff Feedback filter W _fb 。

However, the structure of the active noise reduction earphone results in an objective secondary path in the earphone, and the secondary path may have an adverse effect on the stability of the active noise reduction system in the process of designing the filter. Therefore, how to overcome the influence of the secondary path and improve the system stability is an urgent problem in the art.

Disclosure of Invention

In view of this, the embodiments of the present application provide an active noise reduction method, an active noise reduction device and an active noise reduction earphone, so as to solve the technical problem that the stability of the active noise reduction system in the prior art cannot be further improved due to the influence of the secondary path.

The first aspect of the present application provides an active noise reduction method, including: collecting an original noise signal by a reference microphone; collecting an error signal through an error microphone, wherein the error signal comprises an error between an original noise signal transmitted to a spatial point where the error microphone is located and a noise reduction signal, and the noise reduction signal is determined according to the original noise signal and an initial filter coefficient; inputting the error signal into a non-causal secondary path inverse estimation unit to obtain a pre-estimated error signal, wherein the secondary path inverse estimation unit is used for simulating the inverse of a secondary path; and determining the filter coefficient according to the original noise signal and the estimated error signal.

In an embodiment, the active noise reduction method further includes: a secondary path inverse estimation unit is determined from the secondary path estimation unit, wherein the secondary path estimation unit is configured to simulate a secondary path.

In an embodiment, the active noise reduction method further includes: playing a test signal through a loudspeaker, wherein the test signal is uncorrelated with the original noise signal; obtaining a delay test signal according to the test signal and the initial secondary path estimation unit; determining the difference between the error signal and the delay test signal as a test error signal, wherein the error signal further comprises the test signal transmitted to the spatial point where the error microphone is located; a secondary path estimation unit is determined from the test error signal and the test signal.

In an embodiment, the secondary path inverse estimation unit includes an amplitude estimation unit and a time advance unit, wherein the amplitude estimation unit is used for providing amplitude information of the estimated error signal, and the time advance unit is used for providing phase information of the estimated error signal.

In another embodiment, the secondary path inverse estimation unit comprises a feedback system, wherein an open loop transfer function of the feedback system is determined by a transfer function of the secondary path estimation unit and a gain multiple, the open loop transfer function having a modulus substantially greater than 1.

A second aspect of the present application provides an active noise reduction device comprising: the reference microphone is used for collecting original noise signals; the error microphone is used for collecting error signals, wherein the error signals comprise errors between original noise signals transmitted to space points where the error microphone is located and noise reduction signals, and the noise reduction signals are determined according to the original noise signals and initial filter coefficients; the first calculation module is used for inputting the error signal into a non-causal secondary path inverse estimation unit to obtain a pre-estimated error signal, wherein the secondary path inverse estimation unit is used for simulating the inverse of a secondary path; and the second calculation module is used for determining the filter coefficient according to the original noise signal and the estimated error signal.

A third aspect of the present application provides an active noise reduction earphone comprising: a filter whose filter coefficients are determined by the active noise reduction method provided by any one of the embodiments of the first aspect of the application.

A fourth aspect of the present application provides an active noise reduction earphone comprising: the reference microphone, the loudspeaker, the error microphone and the chip are used for the active noise reduction method provided by any embodiment of the first aspect of the application to determine the filter coefficients in real time.

A fifth aspect of the present application provides an electronic device comprising: a processor; a memory comprising computer instructions stored thereon that, when executed by a processor, cause the processor to perform the active noise reduction method provided by any of the embodiments of the first aspect of the application.

A sixth aspect of the application provides a computer readable storage medium comprising computer instructions stored thereon which, when executed by a processor, cause the processor to perform the active noise reduction method provided by any of the embodiments of the first aspect of the application.

According to the active noise reduction method, the active noise reduction device and the active noise reduction earphone, the error signal (namely the residual noise) is actively advanced in the process of adaptively adjusting the filter coefficient so that the error signal (namely the residual noise) accurately corresponds to the original noise signal (namely the environmental noise) in time, the adverse effect caused by a secondary path can be overcome, and therefore the stability of the active noise reduction process is improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application as claimed.

Drawings

In order to make the objects, technical solutions and advantages of the embodiments of the present application more clear, the embodiments of the present application will be described in further detail below with reference to the accompanying drawings. It is to be understood that the drawings constitute a part of this specification and, together with the examples, serve to explain the application and are not to be taken as limiting the application. In the drawings, like reference numerals and symbols generally refer to like steps or elements unless otherwise indicated.

FIG. 1 is a schematic diagram of an exemplary active noise reduction system.

Fig. 2 is a schematic diagram of an exemplary active noise reduction system according to an embodiment of the present application.

Fig. 3 is a flowchart illustrating an active noise reduction method according to an embodiment of the application.

Fig. 4 is a flowchart illustrating an active noise reduction method according to another embodiment of the application.

Fig. 5 is a schematic diagram of an active noise reduction system according to an embodiment of the application.

Fig. 6 is a schematic diagram of an active noise reduction system according to another embodiment of the application.

Fig. 7 is a schematic diagram of a secondary path measurement system according to an embodiment of the application.

FIG. 8 is a schematic diagram of another exemplary active noise reduction system according to an embodiment of the present application.

Fig. 9 is a flowchart of an active noise reduction method according to another embodiment of the application.

Fig. 10 is a schematic structural diagram of an active noise reduction device according to an embodiment of the application.

Fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the application.

Detailed Description

Application scenario overview

In active noise reduction headphones, the noise reduction parameters (i.e., filter coefficients) of the filter may be determined by way of an off-line design or an on-line design. The offline design means that the filter coefficient is determined before the earphone leaves the factory, and cannot be adjusted again after leaving the factory; the on-line design means that the active noise reduction system in the earphone can adjust the filter coefficient in the use stage of the user so as to make the filter coefficient more fit with the actual noise environment.

In any filter design mode, in the process of determining or adjusting the filter coefficients, the original noise signals collected by the reference microphone and the error noise signals collected by the error microphone can be utilized, the original noise signals and the error noise signals are input into the self-adaptive module, the filter coefficients are gradually adjusted through the self-adaptive calculation process, and the optimal filter coefficients are determined when the error noise signals are converged.

For example, fig. 1 illustrates an active noise reduction system employing an adaptive algorithm, which may employ an LMS (Least Mean Square ) algorithm. The active noise reduction system includes: reference microphone 110, filter 120, speaker (not shown), error microphone 130, and adaptation module 140.

In addition, the paths shown by the broken lines in fig. 1 represent propagation paths of acoustic signals other than the circuit, and specifically include a primary path (transfer function P) formed by the space between the reference microphone 110 to the error microphone 130, and a secondary path (transfer function G) formed by the speaker itself and the space between the speaker and the error microphone 130.

As shown in fig. 1, the original noise at the reference microphone 110 is transferred through the primary path to the point in space where the error microphone 130 is located. Meanwhile, the reference microphone 110 converts the original noise into an original noise signal d (n) after collecting the original noise, and transmits the original noise signal d (n) to the filter 120; the filter 120 calculates a noise reduction signal y (n) having a phase opposite to that of the original noise signal d (n) based on the filter coefficient W from the original noise signal d (n), and outputs it to the speaker; the speaker plays the noise-reduced sound wave based on the noise-reduced signal y (n) such that the noise-reduced sound wave is transferred to the spatial point where the error microphone 130 is located. That is, the noise reduction signal y (n) is output by the filter 120 and then transmitted to the spatial point where the error microphone 130 is located via the secondary path. At this time, the original noise signal d (n) and the noise reduction signal y (n) are transmitted to the error microphone 130 through different paths to form a superposition, so that the error microphone 130 collects an error between the two signals (i.e., the error signal e (n)).

Furthermore, in order to adjust the filter coefficients W, the reference microphone 110 transmits the original noise signal d (n) to the adaptation module 140; the error microphone 130 transmits an error signal e (n) to the adaptation module 140. Based on the original noise signal d (n) and the error signal e (n), the adaptive module 140 iteratively updates the filter coefficients W to finally determine the optimal filter coefficients.

However, as can be appreciated from the above description, unlike the original noise signal d (n) that is directly input from the reference microphone 110 to the adaptation module 140, the error signal e (n) is collected by the error microphone 130 after the noise reduction signal y (n) reaches the spatial point where the error microphone 130 is located through the secondary path. Thus, as shown in fig. 1, the error signal received by the adaptation module 140 simultaneously with the original noise signal d (n) is not e (n) corresponding to d (n), but is an error signal e (n ') corresponding to the original noise signal d (n') whose acquisition time is earlier than d (n). That is, in the prior art, the two input signals that are the basis of the calculation of the adaptive module 140 do not exactly correspond to each other, if the filter coefficients are adjusted according to such input signals, the stability of the active noise reduction system is impaired, and in severe cases, the system may even crash, resulting in that active noise reduction cannot be performed.

In order to solve the problems faced by the existing active noise reduction technology, the embodiment of the application aims to provide an active noise reduction method, an active noise reduction device and an active noise reduction earphone, which are used for improving the stability of an active noise reduction system and further realizing optimized noise reduction by correcting the time difference between two input signals (an original noise signal and an error signal) of an adaptive module.

Exemplary System

Fig. 2 is a schematic diagram of an exemplary active noise reduction system 200 according to an embodiment of the present application. The system comprises: a reference microphone 210, a processor 220, a speaker 230, and an error microphone 240, wherein the processor 220 includes a filter 221, an adaptation module 222, and a secondary path inverse estimation unit 223.

Specifically, the reference microphone 210 is disposed on the earphone housing, and is used for collecting an original noise signal; the filter 221 is configured to receive the original noise signal from the reference microphone 210, calculate a noise reduction signal according to the original noise signal and the filter coefficient, and send the noise reduction signal to the speaker 230; the speaker 230 is configured to play noise-reduced sound waves according to the received noise-reduced signal; the error microphone 240 is disposed near the ear canal of the user, and is used for collecting an error signal (i.e., an error between an original noise signal and a noise reduction signal transmitted to a spatial point where the error microphone 240 is located); the adaptive module 222 is configured to receive the original noise signal from the reference microphone 210 and the error signal from the error microphone 240, and update the filter coefficients according to the original noise signal and the error signal to obtain the optimal filter coefficients.

It should be understood that the paths shown by the dashed lines in fig. 2 represent propagation paths of acoustic signals other than the circuit. Specifically, within the headset, the space between the reference microphone 210 to the error microphone 240 forms a primary path, and the speaker 230 itself and the space between the speaker 230 to the error microphone 240 together form a secondary path. In the active noise reduction process, the original noise is transferred to the error microphone 240 through the primary path, the noise reduction signal is transferred to the error microphone 240 through the secondary path, and the two signals are overlapped at the error microphone 240 to form an error signal.

In addition, as shown in fig. 2, in the active noise reduction system 200 provided in the embodiment of the present application, the processor 220 further includes a secondary path inverse estimation unit 223. Specifically, the secondary path inverse estimation unit 223 is disposed between the error microphone 240 and the adaptive module 222, and when the error microphone 240 sends an error signal to the adaptive module 222, the secondary path inverse estimation unit 223 is configured to process the error signal in advance to obtain an estimated error signal, and input the estimated error signal to the adaptive module 222.

Exemplary method

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application fall within the scope of the present application.

Fig. 3 is a flowchart illustrating an active noise reduction method according to an embodiment of the application. The method may be performed, for example, by the processor 220 in the active noise reduction system 200. As shown in fig. 3, the method includes:

s310: the original noise signal is picked up by a reference microphone.

S320: the error signal is collected by an error microphone.

The error signal comprises an error between an original noise signal transmitted to a spatial point where the error microphone is located and a noise reduction signal, and the noise reduction signal is determined according to the original noise signal and an initial filter coefficient.

Specifically, after the reference microphone collects the original noise signal d (n) at the first time, the filter may calculate the noise reduction signal y (n) corresponding to d (n) based on d (n) and the initial filter coefficient (i.e., the filter coefficient to be adjusted), and output the noise reduction signal y (n) to the secondary path. y (n) passes through the secondary path to reach the spatial point where the error microphone is located, so that the error microphone collects error signals e (n) corresponding to d (n) and y (n) at a second moment.

In this process, the second moment when the error microphone collects the error signal e (n) is necessarily later than the first moment when the reference microphone collects the original noise signal d (n), subject to the delay of the secondary path.

S330: and inputting the error signal into a non-causal secondary path inverse estimation unit to obtain an estimated error signal.

Wherein the secondary path inverse estimation unit is used for simulating the inverse of the secondary path.

At a first instant in time, the error signal acquired by the error microphone is not the error signal e (n) corresponding to d (n), but the error signal e (n ') corresponding to the original noise signal d (n') acquired earlier than d (n). At this time, if the error signal e (n ') is directly input to the adaptation module, the adaptation module will calculate based on e (n') and d (n) which do not correspond to each other, resulting in poor system stability. As previously mentioned, this phenomenon is caused by the delay caused by the secondary path.

Therefore, if e (n) that could not be acquired at the second time can be obtained in advance at the first time based on e (n') acquired at the first time, e (n) and the original noise signal d (n) can be input to the adaptive module at the first time, so that the adaptive module receives d (n) and e (n) that correspond to each other in time at the same time.

Specifically, in an embodiment of the present invention, a secondary path inverse estimation unit may be provided between the error microphone and the adaptation module. The secondary path inverse estimation unit is used to simulate the inverse of the secondary path, and "cancel" the effect of the secondary path. After receiving the error signal e (n ') acquired by the error microphone at the first moment, the secondary path inverse estimation unit may perform advanced processing on the error signal e (n'), that is, predict in advance to obtain a predicted error signal e '(n) for simulating the real error signal e (n), and input the predicted error signal e' (n) into the adaptive module at the first moment, so that the error signal and the original noise signal are mutually aligned in time, thereby eliminating the influence caused by the secondary path.

S340: and determining the filter coefficient according to the original noise signal and the estimated error signal.

After receiving the original noise signal and the estimated error signal, the adaptive module can adjust the initial filter coefficient, judge whether the adjusted filter coefficient is the optimal filter coefficient, if not, adjust again, repeat the above process until the filter coefficient is optimal.

Specifically, in one embodiment, it may be determined whether the filter coefficients are optimal based on the estimated error signal.

When the estimated error signal is judged to not reach the preset optimal condition, the initial filter coefficient can be adjusted, and the updated noise reduction signal is determined by adopting the adjusted filter coefficient. After the updated noise reduction signal is played by the loudspeaker, the error microphone can acquire the updated error signal, so that the self-adaptive module obtains the updated estimated error signal. And when the updated estimated error signal still does not reach the preset optimal condition, the filter coefficient can be adjusted again to obtain the updated estimated error signal again. And repeating the steps until the estimated error signal meets the preset optimal condition, stopping adjustment, and determining the current filter coefficient (namely the filter coefficient after the last adjustment) as the final filter coefficient.

In an embodiment, for example, the energy of the estimated error signal reaches a minimum value, that is, whether the estimated error signal meets a preset optimal condition is determined by determining whether the energy of the residual noise signal reaches a minimum value.

Here, the process of iteratively adjusting the filter coefficients and updating the estimated error signal may be implemented using an adaptive algorithm, such as an LMS (Least Mean Square ) algorithm, each time the filter coefficients are updated until the estimated error signal is optimal. It should be understood that embodiments of the present application are not limited to the actual algorithm employed.

According to the active noise reduction method provided by the embodiment of the application, the error signal is actively advanced in the process of adaptively adjusting the filter coefficient so as to accurately correspond to the original noise signal in time, so that the adverse effect caused by the secondary path can be overcome, and the stability of the active noise reduction process is improved.

Fig. 4 is a flowchart illustrating an active noise reduction method according to another embodiment of the application. The method may be performed, for example, by the processor 220 in the active noise reduction system 200. As shown in fig. 4, on the basis of the method shown in fig. 3, the method further includes:

S410: a secondary path inverse estimation unit is determined from the secondary path estimation unit.

Wherein the secondary path estimating unit is used for simulating a secondary path.

It should be appreciated that the secondary path (transfer function G) is a causal system, and thus its inverse is a non-causal system. That is, the secondary path inverse estimation unit is non-causal and cannot be directly implemented by physical means. Thus, in an embodiment of the present application, the secondary path estimating unit may be determined first by physical means, and then the secondary path inverse estimating unit may be determined further based on the secondary path estimating unit.

Here, since the secondary path estimating unit may be predetermined or may be adjusted in real time during the use of the earphone, S410 may be implemented before S310 or after S320, which is not limited by the embodiment of the present application.

In an embodiment, the manner of determining the secondary path inverse estimation unit according to the secondary path estimation unit may be as shown in fig. 5.

In the active noise reduction system 500 shown in fig. 5, the secondary path inverse estimation unit 523 may include an amplitude estimation unit 510 and a time advance unit 520. The amplitude estimation unit 510 is obtained according to the secondary path estimation unit, and is configured to provide amplitude information of the estimated error signal; the time advance unit 520 is used for providing the phase information of the estimated error signal.

Specifically, the amplitude estimation unit 510 is equivalent to a filter, and can process the error signal collected by the error microphone based on its own parameters to obtain the amplitude information of the estimated error signal. Here, the parameters of the amplitude estimation unit 510 may be determined according to the secondary path estimation unit. That is, after the change condition of the secondary path to the amplitude of the error signal is determined by the secondary path estimating unit, the changed amplitude can be modified to the amplitude before the change by the pushback. It should be appreciated that the amplitude estimation unit 510 can be physically implemented, for example, based on the secondary path estimation unit, by designing the amplitude estimation unit 510 by designing an FIR filter.

The time advance unit 520 can modify the phase of the error signal, and infer the phase information of the error signal to be collected at the second time at the first time.

After the amplitude information and the phase information are provided, the secondary path inverse estimation unit can obtain an estimated error signal according to the received error signal.

In another embodiment, the manner of determining the secondary path inverse estimation unit from the secondary path estimation unit may be as shown in fig. 6.

In the active noise reduction system 600 shown in fig. 6, the secondary path inverse estimation unit 623 may be set as the feedback system 610, wherein an open loop transfer function of the feedback system is determined by a gain multiple and a transfer function of the secondary path estimation unit, and a modulus of the open loop transfer function is much greater than 1.

In particular, the transfer function of the secondary path estimation unit may be denoted as G ', the gain multiple may be denoted as K, the open loop transfer function of the feedback system 610 may be denoted as KG ', and the closed loop transfer function may be denoted as K/(1+kg '). Thus, when the modulus of the open loop transfer function is much greater than 1, the closed loop transfer function of the feedback system 610 is infinitely close to 1/G', i.e. the inverse of the secondary path estimation unit.

The active noise reduction method provided by the embodiment of the application can obtain the non-causal secondary path inverse estimation unit based on the physically-realizable secondary path estimation unit, the technical scheme is easy to realize, a large amount of computing resources are not required to be consumed, and a better solution is provided for the field.

It should be noted that, in some embodiments of the present application, the secondary path estimating unit may be predetermined by offline calculation; in another part of embodiments of the present application, the secondary path estimation unit may also adjust the optimization in real time through online calculation during actual use by the user, so as to adapt to different use environments.

For example, in one embodiment of the present application, the secondary path estimation unit may be predetermined by means of off-line calculation using the secondary path measurement system 700 shown in fig. 7. As shown in fig. 7, the secondary path measurement system 700 includes a player 710, a processor 720, a speaker 730, and an error microphone 740. The processor 720 includes a secondary path estimation unit 721, an adder 722, and an adaptation module 723.

The method of determining the secondary path estimation unit based on the secondary path measurement system 700 may be implemented in a filter design stage prior to shipment of the headset, and in particular may be performed by the processor 720 in the secondary path measurement system 700 shown in fig. 7. The method may comprise the steps of:

playing the test signal through a loudspeaker;

obtaining a delay test signal according to the test signal and the initial secondary path estimation unit;

determining a test error signal based on the delayed test signal and the test signal communicated to the error microphone;

a secondary path estimation unit is determined from the test error signal and the test signal.

Specifically, the test signal may be a white noise signal, a pink noise signal, etc. from the player 710, and the embodiment of the present application does not limit the specific selection of the test signal.

The player 710 inputs a test signal to the speaker 730 via a circuit, and the test signal is collected by the error microphone 740 after being transferred from the speaker 730 to a spatial point where the error microphone 740 is located, i.e., after passing through a secondary path. That is, the error microphone 740 collects the test signal affected by the delay of the secondary path.

Specifically, as shown in fig. 7, in the secondary path measurement system 700, a secondary path estimation unit 721 is provided between the player 710 and the adder 722. Here, the unadjusted secondary path estimating unit is an initial secondary path estimating unit.

The player 710 inputs the test signal to the initial secondary path estimation unit via the circuit, and after the initial secondary path estimation unit receives the test signal, the test signal may be delayed to obtain a delayed test signal, and input it to the adder 722.

Specifically, error microphone 740 may input a test signal after receiving the secondary path delay affected signal to adder 722. Adder 722 may compare the delayed test signal with the test signal after the secondary path delay effect to obtain an error therebetween, i.e., a test error signal.

It should be appreciated that the energy of the test error signal should be minimized when the secondary path estimation unit 721 is infinitely close to the true secondary path, where the secondary path estimation unit is the optimal secondary path estimation unit.

Therefore, to obtain an optimal secondary path estimation unit, the test signal and the test error signal may be input to the adaptive module 723, and the initial secondary path estimation unit may be iteratively adjusted by an adaptive algorithm to obtain an updated test error signal. When it is determined that the expected power of the current updated test error signal reaches a minimum value, then the current (after the last adjustment) secondary path estimation unit may be determined to be the closest secondary path estimation unit to the actual secondary path.

According to the active noise reduction method provided by the embodiment of the application, the secondary path estimation unit is obtained through self-adaptive adjustment in the filter design stage, so that the secondary path inverse estimation unit can be determined, the stability of the system is improved by utilizing the secondary path inverse estimation unit in the active noise reduction process, and good use experience is brought to users.

In other embodiments, other methods may be used to perform the offline calculation of the secondary path estimation unit. For example, playing a white noise signal x (n) through a loudspeaker, and collecting the white noise signal x (n) through an error microphone to obtain y (n); solving the self-power spectrum P of x (n) respectively _xx And x (n) and y (n) cross power spectrum P _xy The method comprises the steps of carrying out a first treatment on the surface of the According to P _xx And P _xy And calculating a transfer function of the secondary path, and setting a secondary path estimating unit based on the transfer function of the secondary path.

Fig. 8 is a schematic diagram of another exemplary active noise reduction system 800 according to an embodiment of the present application. The system differs from the exemplary active noise reduction system 200 shown in fig. 2 in that it further comprises a player 810, while the processor 820 comprises a filter 221, a first adaptation module 222, a secondary path inverse estimation unit 823, a secondary path estimation unit 824, an adder 825 and a second adaptation module 826.

Fig. 9 is a flowchart of an active noise reduction method according to another embodiment of the application. The method may be performed, for example, by processor 820 in the active noise reduction system 800 shown in fig. 8.

As previously described, fig. 7 of the present application shows a system for off-line calculation of a secondary path estimation unit. However, since the secondary path estimating unit obtained by offline calculation is predetermined before the earphone leaves the factory, it cannot be changed during actual use. Therefore, in actual use of the user, the calibrated secondary path estimation unit cannot adapt to the actual secondary path of each earphone in real time, and the system cannot determine a more effective secondary path inverse estimation unit according to the actual secondary path, so that user experience cannot reach the best.

In view of this, the embodiment of the present application provides the active noise reduction method shown in fig. 9, which can implement on-line calculation of the secondary path estimating unit and update the secondary path inverse estimating unit in real time. For example, during actual use by a user, processor 820 may perform the method to optimize the secondary path inverse estimation unit in real time and employ the optimized secondary path inverse estimation unit for active noise reduction.

It should be understood that, according to practical needs, the method shown in fig. 9 may also be used for calculating the secondary path estimating unit offline, and the embodiment of the present application is not limited to the specific application scenario of the method.

As shown in fig. 9, the online computing method includes:

s910: the original noise signal is picked up by a reference microphone.

S920: the test signal is played through a speaker.

Wherein the test signal is a signal that is uncorrelated with the original noise signal. In particular, the test signal may be a swept frequency signal from the player 810, or the like.

Preferably, in another implementation manner, since the active noise reduction method provided in this embodiment may be performed in actual use by a user, the test signal may also be an acoustic signal actually played by the user, such as a voice call signal, a media audio signal, or the like. Such signals are non-stationary signals (mainly medium and high frequencies) and can be regarded as having no correlation with the original noise signal from the environment (stationary, mainly low frequencies).

It should be understood that S920 may also be performed before S910.

S930: the error signal is collected by an error microphone.

Similar to the previous embodiment, the filter 221 generates a noise reduction signal from the original noise signal and the filter coefficients and transmits it to the speaker 230 for playback. The original noise signal and the noise reduction signal respectively pass through the primary path and the secondary path and reach the spatial point where the error microphone 240 is located, so that the error microphone 240 collects the error between the two.

In addition, in the present embodiment, the test signal also passes through the secondary path to reach the spatial point where the error microphone 240 is located. Since the test signal is uncorrelated with the original noise signal, it is understood that the error signal collected by the error microphone 240 includes the error between the original noise signal and the noise reduction signal that are transmitted to the spatial point where the error microphone 240 is located, and also includes the test signal.

S940: and obtaining a delay test signal according to the test signal and the initial secondary path estimation unit.

Similar to the embodiment shown in fig. 7, in this embodiment, a secondary path estimating unit 824 for simulating a secondary path may be provided between the player 810 and the adder 825. Here, the unadjusted secondary path estimating unit is an initial secondary path estimating unit.

The player 810 may directly send the test signal to the initial secondary path estimation unit through the circuit, and after the initial secondary path estimation unit receives the test signal, the test signal may be delayed to obtain a delayed test signal, and send the delayed test signal to the adder 825.

In another embodiment, the secondary path estimation unit 824 may also be disposed between the speaker 230 and the adder 825 (not shown). At this time, the speaker 230 may input the noise reduction signal and the test signal to the initial secondary path estimating unit through a circuit while playing the noise reduction signal and the test signal after receiving the noise reduction signal from the filter 221 and the test signal from the player 810. After the initial secondary path estimation unit receives the noise reduction signal and the test signal, delay processing can be performed to obtain a delay test signal. It should be appreciated that in this embodiment, the delayed test signal includes the noise reduction signal and the test signal which are delayed.

S950: the difference between the error signal and the delayed test signal is determined as the test error signal.

Specifically, the error microphone 240, upon receiving the error signal, may input it to an adder 825. Adder 825 may compare the error signal from error microphone 240 with the delayed test signal from the initial secondary path estimation unit to derive an error therebetween, i.e., a test error signal.

S960: a secondary path estimation unit is determined from the test error signal and the test signal.

Similar to the previous embodiment, the energy of the test error signal should be minimized when the secondary path estimation unit 824 is infinitely close to the real secondary path, which is the optimal secondary path estimation unit.

Thus, to obtain an optimal secondary path estimation unit, the test signal and the test error signal may be input to the second adaptation module 826, and the initial secondary path estimation unit may be iteratively adjusted by an adaptation algorithm to obtain an updated test error signal. When it is determined that the expected power of the current updated test error signal reaches a minimum value, then the current (after the last adjustment) secondary path estimation unit may be determined to be the closest secondary path estimation unit to the actual secondary path.

S970: a secondary path inverse estimation unit is determined from the secondary path estimation unit.

As previously mentioned, the secondary path inverse estimation unit is non-causal for modeling the inverse of the secondary path. The specific manner of determining the secondary path inverse estimation unit according to the secondary path estimation unit may refer to the embodiments shown in fig. 4 to 6, and will not be described herein.

S980: and inputting the error signal into a secondary path inverse estimation unit to obtain an estimated error signal.

Similarly to the embodiment shown in fig. 3, to overcome the time difference between the original noise signal and the error signal, the error signal may be first processed in advance by the secondary path inverse estimation unit to obtain a corresponding estimated error signal, and input to the first adaptive module 222.

It should be appreciated that in the active noise reduction system 800, after the secondary path estimation unit 824 is determined by the second adaptation module 826, the secondary path inverse estimation unit 823 may be updated according to the secondary path estimation unit, and the error signal may be processed in advance using the updated secondary path inverse estimation unit 823 to obtain the estimated error signal, as shown in the dashed line portion of fig. 8.

S990: and determining the filter coefficient according to the original noise signal and the estimated error signal.

Similar to the embodiment shown in fig. 3, after the first adaptive module 222 receives the original noise signal and the estimated error signal, the initial filter coefficient may be adjusted, and it may be determined whether the adjusted filter coefficient is an optimal filter coefficient, if not, the adjustment is performed again, and the above process is repeated until the filter coefficient reaches the optimal filter coefficient.

Specifically, in one embodiment, it may be determined whether the filter coefficients are optimal based on the estimated error signal. When the estimated error signal is judged to not reach the preset optimal condition, the initial filter coefficient can be adjusted, and the updated noise reduction signal is determined by adopting the adjusted filter coefficient. After the speaker 230 plays the updated noise reduction signal, the error microphone 240 can collect the updated error signal, so that the first adaptive module 222 obtains the updated estimated error signal. And when the updated estimated error signal still does not reach the preset optimal condition, the filter coefficient can be adjusted again to obtain the updated estimated error signal again. And repeating the steps until the estimated error signal meets the preset optimal condition, stopping adjustment, and determining the current filter coefficient as the final filter coefficient.

Here, the process of iteratively adjusting the filter coefficients and updating the estimated error signal may be implemented using an adaptive algorithm, such as an LMS algorithm, each time the filter coefficients are updated until the estimated error signal is optimal. It should be understood that embodiments of the present application are not limited to the actual algorithm employed.

In an embodiment, S960 specifically may include the following steps:

adjusting the initial secondary path estimation unit according to the test error signal;

a. determining an updated test error signal based on the test signal and the adjusted secondary path estimation unit;

b. when the expected power of the updated test error signal does not reach the minimum value, the adjusted secondary path estimation unit is adjusted;

iteratively executing the steps a and b until the expected power of the test error signal reaches the minimum value;

and determining the current adjusted secondary path estimation unit as a secondary path estimation unit.

Here, the process of iteratively performing steps a, b until the expected power of the updated test error signal reaches a minimum may be implemented using an adaptive algorithm, such as an LMS algorithm.

It should be understood that in other embodiments of the present application, some steps of the method shown in fig. 9 may be omitted, or may be performed in other sequences.

For example, in one embodiment, the above method may be performed without turning on the active noise reduction function, i.e., S980, S990 may be omitted, and filters and noise reduction signals are not required during the execution of each step. That is, when the active noise reduction is not started by the user, the secondary path inverse estimation unit can be debugged only by using the original noise signal and the test signal, and the filter coefficient is not adjusted temporarily. Therefore, when the user opens the noise reduction function, the secondary path inverse estimation unit is in an optimal state, and can directly enter a link for optimizing the filter coefficient, so that a large amount of debugging time can be saved.

For another example, in another embodiment, the filter coefficients may be iteratively adjusted while the secondary path estimation unit is iteratively adjusted. That is, before determining the secondary path estimating unit and thus determining the secondary path inverse estimating unit, that is, when the secondary path inverse estimating unit has not reached the optimum, the current secondary path inverse estimating unit may be dynamically determined according to the current adjusted secondary path estimating unit after each adjustment of the secondary path estimating unit, and used in the adaptive link of the filter coefficient, the error signal is processed in advance with the current adjusted secondary path inverse estimating unit and input to the first adaptive module 222. Through the execution sequence, the secondary path inverse estimation unit and the adjustment operation of the filter coefficient can be synchronously performed, so that the optimization of the filter coefficient can be started as soon as possible, and the user experience is improved.

According to the active noise reduction method provided by the embodiment of the application, by adopting the sound signal actually played by the user as the test signal, the optimization of the secondary path inverse estimation unit can be naturally realized when the user plays the audio or calls by using the earphone, the stability of the active noise reduction system is improved, the debugging process is more friendly and softer, and great convenience and good experience are brought to the user.

In real life, the degree of perception of sounds of different frequencies by the human ear is different. Therefore, the sound pressure level of the real noise collected by the microphone and the noise heard by the human ear at the same frequency are not necessarily the same. For active noise reduction headphones, the primary purpose is to reduce the noise level heard by the human ear, and not simply to focus on the actual noise level.

From this point of view, the active noise reduction method provided by the other embodiment of the present application can adaptively update the filter coefficient on the basis of considering the response characteristics of the human ear to noise with different frequencies, so as to achieve the active noise reduction effect of more fitting the human ear.

In an embodiment, on the basis of the active noise reduction system shown in fig. 2 or fig. 8, the processor may further include an acoustic weighting unit disposed between the error microphone and the adaptive module (or the first adaptive module). It should be understood that the acoustic weighting unit may be disposed between the error microphone and the secondary path inverse estimation unit, or may be disposed between the adaptive module (or the first adaptive module) and the secondary path inverse estimation unit, which is not limited by the embodiment of the present application.

In this embodiment, before the error signal is input to the secondary path inverse estimation unit, the error signal may be input to the acoustic weighting unit to obtain a weighted error signal, and then the weighted error signal is input to the secondary path inverse estimation unit to calculate a weighted estimated error signal; alternatively, after obtaining the estimated error signal according to the error signal and the secondary path inverse estimation unit, the estimated error signal may be input into the acoustic weighting unit to obtain the weighted estimated error signal.

Specifically, the weighted error signal can be obtained by performing weighted correction on the spectrum shape of the error signal or the estimated error signal through the acoustic weighting unit. For example, the acoustic weighting unit may perform a-weighting on the error signal so that the noise spectrum of the weighted estimated error signal is more similar to the auditory sense of the human ear.

According to the active noise reduction method provided by the embodiment of the application, the weighted estimated error signal is input into the self-adaptive module to participate in the adjustment process of the filter coefficient instead of the unprocessed estimated error signal, so that the filter coefficient can be guided to optimize towards the actual noise reduction requirement direction of human ears, the active noise reduction effect is obviously improved, and better hearing experience is brought to users.

It should be noted that, the embodiment of the present application does not limit whether the earphone is further provided with a feedback noise reduction system, and the feedback noise reduction loop may be added to the active noise reduction system provided in any of the above embodiments of the present application to form a hybrid active noise reduction earphone including a feedforward adaptive active noise reduction system and a feedback active noise reduction system.

Exemplary apparatus

Fig. 10 is a schematic structural diagram of an active noise reduction device 1000 according to an embodiment of the application.

As shown in fig. 10, the active noise reduction device 1000 includes: a reference microphone 1010 for collecting an original noise signal; the error microphone 1020 is configured to collect an error signal, where the error signal includes an error between an original noise signal transmitted to a spatial point where the error microphone 1020 is located and a noise reduction signal, and the noise reduction signal is determined according to the original noise signal and an initial filter coefficient; a first calculation module 1030, configured to input the error signal into a non-causal secondary path inverse estimation unit, to obtain an estimated error signal, where the secondary path inverse estimation unit is configured to simulate an inverse of a secondary path; a second calculation module 1040 is configured to determine filter coefficients according to the original noise signal and the estimated error signal.

Specifically, in order to enable the second calculation module 1040 to perform the adjustment operation of the filter coefficient based on the original noise signal and the error signal that accurately correspond to each other, the first calculation module 1030 provided in the embodiment of the present application is provided with a secondary path inverse estimation unit for simulating the inverse of the secondary path. After the first calculation module 1030 receives the error signal collected by the error microphone, the error signal may be processed in advance by using the inverse estimation unit of the secondary path, to obtain an estimated error signal corresponding to the original noise, and the estimated error signal is input into the second calculation module 1040, so that the error signal (i.e. the estimated error signal) received by the second calculation module 1040 and the original noise signal are mutually aligned in time, thereby canceling the influence caused by the secondary path.

The second calculation module 1040 may include an adaptive module, which may adjust the initial filter coefficient after receiving the original noise signal and the estimated error signal, and determine whether the adjusted filter coefficient is an optimal filter coefficient, and if not, adjust again, and repeat the above process until the filter coefficient reaches the optimal filter coefficient.

Specifically, in an embodiment, the second calculating module 1040 may include an adaptive module, and the adaptive module may determine whether the filter coefficient is optimal based on the estimated error signal.

When the estimated error signal is judged to not reach the preset optimal condition, the initial filter coefficient can be adjusted, and the updated estimated noise reduction signal is determined by adopting the adjusted filter coefficient. After the speaker plays the updated noise reduction signal, the error microphone 1020 can acquire the updated error signal, and the secondary path inverse estimation unit processes the updated error signal to obtain an updated estimated error signal. When the updated estimated error signal still does not reach the preset optimal condition, the adaptive module can adjust the filter coefficient again to obtain the updated estimated error signal again. And repeating the steps until the estimated error signal meets the preset optimal condition, stopping adjustment, and determining the current filter coefficient (namely the filter coefficient after the last adjustment) as the final filter coefficient.

According to the active noise reduction device provided by the embodiment of the application, the error signal is actively advanced in the process of adaptively adjusting the filter coefficient so as to accurately correspond to the original noise signal in time, so that the adverse effect caused by the secondary path can be overcome, and the stability of the active noise reduction process is improved.

In an embodiment, the first calculation module 1030 is further configured to determine a secondary path inverse estimation unit according to the secondary path estimation unit. Wherein the secondary path estimating unit is used for simulating a secondary path.

Since the secondary path inverse estimation unit is non-causal and cannot be directly implemented by physical means, in this embodiment, the secondary path estimation unit may be determined first, and then the secondary path inverse estimation unit may be further determined based on the secondary path estimation unit.

The manner in which the secondary path inverse estimation unit is determined according to the secondary path estimation unit may refer to the embodiments described in fig. 4 to 6 in the exemplary method, and will not be described in detail herein.

It should be noted that, in some embodiments of the present application, the secondary path estimation unit may be predetermined through offline calculation, and in other embodiments of the present application, the secondary path estimation unit may also adjust and optimize through online calculation in real time during actual use of the user, so as to adapt to different use environments.

The manner in which the secondary path estimating unit is predetermined by the offline computing method may refer to the related content of the embodiment shown in fig. 7 in the exemplary method, which is not described herein.

An embodiment of the present application provides an active noise reduction device capable of adjusting a secondary path inverse estimation unit through online calculation, and the active noise reduction device further includes a third calculation module on the basis of the device shown in fig. 10.

Specifically, when the active noise reduction device provided by the embodiment performs active noise reduction, a test signal from the player can be played by using a speaker in the earphone, wherein the test signal is not related to the original noise signal.

Here, the third calculation module includes a secondary path estimation unit. The third calculation module can receive the test signal from the player through the circuit, and further obtains a delay test signal according to the test signal and an initial secondary path estimation unit (the unadjusted secondary path estimation unit is the initial secondary path estimation unit). Or the third calculation module can also receive the noise reduction signal and the test signal from the loudspeaker through the circuit, and simultaneously process the two signals based on the initial secondary path estimation unit to obtain the processed noise reduction signal and the test signal to be used as the delay test signal.

Further, the third calculation module may receive the error signal from the error microphone and determine a difference between the error signal and the delayed test signal as the test error signal. It should be appreciated that in this embodiment, the error signal includes both the error between the original noise signal and the noise reduction signal that are transmitted to the point of space where the error microphone is located, and the test signal that is transmitted to the point of space where the error microphone is located.

It will be appreciated that the energy of the test error signal should be minimized when the secondary path estimating unit is infinitely close to the true secondary path, the secondary path estimating unit being the optimal secondary path estimating unit.

Therefore, in order to obtain the optimal secondary path estimation unit, the third calculation module may repeatedly adjust the initial secondary path estimation unit according to the test error signal and the test signal through an adaptive algorithm, so as to obtain an updated test error signal. When it is determined that the expected power of the current updated test error signal reaches a minimum value, then the current (after the last adjustment) secondary path estimation unit may be determined to be the closest secondary path estimation unit to the actual secondary path.

After determining the secondary path estimating unit, the third computing module may synchronize the secondary path estimating unit to the first computing module such that the first computing module determines an updated secondary path inverse estimating unit based on the updated secondary path estimating unit and adjusts the filter coefficients using the updated secondary path inverse estimating unit.

It should be appreciated that in other embodiments of the present application, the above-described partial construction of the active noise reduction device may be used to actively reduce noise in other ways or sequences.

For example, in one embodiment, the above-described method may be performed without turning on the active noise reduction function, i.e., the filters and noise reduction signals may not be used during the execution of the steps. That is, when the active noise reduction is not started by the user, the active noise reduction device may only use the original noise signal and the test signal to debug the secondary path inverse estimation unit, and the filter coefficient is not adjusted temporarily. Therefore, when the user opens the noise reduction function, the secondary path inverse estimation unit is in an optimal state, and can directly enter a link for optimizing the filter coefficient, so that a large amount of debugging time can be saved.

For another example, in another embodiment, the filter coefficients may be iteratively adjusted while the secondary path estimation unit is iteratively adjusted. That is, before determining the secondary path estimating unit, i.e., when the secondary path estimating unit has not reached the optimum, the third calculating module may dynamically synchronize the current adjusted secondary path estimating unit to the first calculating module after each adjustment of the secondary path estimating unit, so that the first calculating module updates the secondary path inverse estimating unit based on the current adjusted secondary path estimating unit and uses it in the adaptation link of the filter coefficients. Through the execution sequence, the secondary path inverse estimation unit and the adjustment operation of the filter coefficient can be synchronously performed, so that the optimization of the filter coefficient can be started as soon as possible, and the user experience is improved.

According to the active noise reduction device provided by the embodiment of the application, by adopting the sound signal actually played by the user as the test signal, the optimization of the secondary path inverse estimation unit can be naturally realized when the user plays the audio or calls by using the earphone, the stability of the active noise reduction system is improved, the debugging process is more friendly and softer, and great convenience and good experience are brought to the user.

Further, in another embodiment, the first computing module may further include an acoustic weighting unit. In the active noise reduction process, the acoustic weighting unit can process an error signal (the acoustic weighting unit is arranged between the error microphone and the secondary path inverse estimation unit) or an estimated error signal (the acoustic weighting unit is arranged between the secondary path inverse estimation unit and the self-adaptive module), so as to obtain a weighted error signal or an estimated error signal. In this way, the weighted estimated error signal is input to the second calculation module for adjusting the filter coefficients instead of the unweighted estimated error signal.

Specifically, the acoustic weighting unit may perform weighted correction on the spectrum shape of the error signal or the estimated error signal, to obtain a weighted error signal or an estimated error signal. For example, the acoustic weighting unit may perform a-weighting so that the noise spectrum of the weighted error signal or the estimated error signal is more similar to the hearing of the human ear.

According to the active noise reduction device provided by the embodiment of the application, the weighted estimated error signal is used as an adaptively calculated input signal to participate in the adjustment process of the filter coefficient, so that the filter coefficient can be guided to optimize towards the actual noise reduction requirement direction of human ears, the active noise reduction effect is obviously improved, and better hearing experience is brought to users.

It should be understood that the functions and technical effects of each module in the active noise reduction device provided in the foregoing embodiments may refer to corresponding content in the exemplary method, which is not described herein in detail.

Exemplary apparatus

Fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the application. As shown in fig. 11, the electronic device includes: a processor 1110; memory 1120, memory 1120 including computer instructions stored thereon that, when executed by processor 1110, cause processor 1110 to perform the active noise reduction method provided by any of the embodiments described above.

Exemplary computer-readable storage Medium

Other embodiments of the present application also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the active noise reduction method according to any of the embodiments described above. It is understood that the computer storage medium may be any tangible medium, such as: floppy disks, CD-ROMs, DVDs, hard drives, or network media.

The block diagrams of the devices, apparatus, systems according to the present application are merely illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. Those skilled in the art will appreciate that the devices, apparatuses, systems may be connected, arranged, and configured in any manner. Words such as "comprising," "including," "having," and the like are open ended terms to "including, but not limited to," and are used interchangeably herein, unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.

It is also noted that in the apparatus, devices and methods of the present application, the modules or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown above but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing description is provided for the purpose of illustration and description of the application. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed above. Although a number of example aspects and embodiments have been discussed in the foregoing, other variations, modifications, changes, additions, and sub-combinations will readily occur to those skilled in the art based upon the foregoing.

The foregoing description of the preferred embodiments of the application is not intended to be limiting, but rather is to be construed as including any modifications, equivalents, and alternatives falling within the spirit and principles of the application.

Claims

1. An active noise reduction method, comprising:

collecting an original noise signal by a reference microphone;

collecting an error signal through an error microphone, wherein the error signal comprises an error between an original noise signal transmitted to a spatial point where the error microphone is located and a noise reduction signal, and the noise reduction signal is determined according to the original noise signal and an initial filter coefficient;

inputting the error signal into a non-causal secondary path inverse estimation unit to obtain a pre-estimated error signal, wherein the secondary path inverse estimation unit is used for simulating the inverse of a secondary path;

according to the original noise signal and the estimated error signal, adjusting the initial filter coefficient, and determining an updated noise reduction signal by adopting the adjusted filter coefficient to obtain an updated estimated error signal;

judging whether the energy of the updated estimated error signal reaches the minimum value, if not, adjusting the adjusted filter coefficient again to obtain a re-updated estimated error signal;

repeating the above steps until the energy of the finally updated estimated error signal reaches the minimum value, determining the current adjusted filter coefficient as the filter coefficient,

Wherein the secondary path inverse estimation unit is determined from a secondary path estimation unit for simulating the secondary path,

the secondary path inverse estimation unit comprises an amplitude estimation unit and a time advance unit, wherein the amplitude estimation unit is used for providing amplitude information of the estimated error signal, and the time advance unit is used for providing phase information of the estimated error signal; alternatively, the secondary path inverse estimation unit comprises a feedback system, an open loop transfer function of which is determined by a transfer function of the secondary path estimation unit and a gain multiple, the open loop transfer function having a modulus substantially greater than 1.

2. The active noise reduction method of claim 1, further comprising:

playing a test signal through a loudspeaker, wherein the test signal is uncorrelated with the original noise signal;

obtaining a delay test signal according to the test signal and an initial secondary path estimation unit;

determining the difference between the error signal and the delay test signal as a test error signal, wherein the error signal further comprises the test signal transmitted to the spatial point where the error microphone is located;

And determining the secondary path estimation unit according to the test error signal and the test signal.

3. An active noise reduction device, comprising:

the reference microphone is used for collecting original noise signals;

the error microphone is used for collecting error signals, wherein the error signals comprise errors between original noise signals transmitted to space points where the error microphones are located and noise reduction signals, and the noise reduction signals are determined according to the original noise signals and initial filter coefficients;

the first calculation module is used for inputting the error signal into a non-causal secondary path inverse estimation unit to obtain a pre-estimated error signal, wherein the secondary path inverse estimation unit is used for simulating the inverse of a secondary path;

a second calculation module for determining a filter coefficient based on the original noise signal and the pre-estimated error signal,

wherein the second computing module comprises an adaptation module for: according to the original noise signal and the estimated error signal, adjusting the initial filter coefficient, and determining an updated noise reduction signal by adopting the adjusted filter coefficient to obtain an updated estimated error signal; judging whether the energy of the updated estimated error signal reaches the minimum value, if not, adjusting the adjusted filter coefficient again to obtain a re-updated estimated error signal; repeating the above steps until the energy of the finally updated estimated error signal reaches the minimum value, determining the current adjusted filter coefficient as the filter coefficient,

4. An active noise reduction earphone, comprising:

a filter whose filter coefficients are determined by the active noise reduction method of any one of claims 1-2.

5. An active noise reduction earphone, comprising: a reference microphone, a speaker, an error microphone, and a chip,

wherein the reference microphone, speaker, error microphone and chip are adapted to perform the active noise reduction method of any of claims 1-2 to determine the filter coefficients in real time.

6. An electronic device, comprising:

a processor;

a memory comprising computer instructions stored thereon that, when executed by the processor, cause the processor to perform the active noise reduction method of any of claims 1-2.

7. A computer readable storage medium comprising computer instructions stored thereon, which when executed by a processor, cause the processor to perform the active noise reduction method of any of claims 1-2.