Echo cancellation method, device and communication equipment

Abstract

The present invention relates to an echo elimination method, device and communication equipment. A reference microphone arranged close to a loudspeaker is used to receive a second signal, a vibration sensor arranged close to a receiving microphone is used to receive a third signal, a nonlinear signal is obtained according to a first signal, a far-end signal and a second signal, and a dual-channel reference signal combining the nonlinear characteristics of the loudspeaker and a device vibration signal caused by the loudspeaker is obtained by splicing the nonlinear signal and the third signal. A propagation path of a signal is obtained based on the dual-channel reference signal and the first signal, and an echo interference signal received during the transmission of the far-end signal is obtained by convolving the dual-channel reference signal and the equivalent propagation path. The echo interference signal is eliminated from the first signal, so that the problem of incomplete echo elimination caused by the reception distortion caused by the loudspeaker distortion and the vibration of the receiving microphone is avoided, thereby improving the audio transmission quality.

Description

Technical Field

[0001] The present invention relates to the field of echo cancellation, and in particular to an echo cancellation method, device and communication equipment. Background Art

[0002] In communication systems, loudspeakers are commonly used to generate sound pressure waves and broadcast them, while microphones receive the sound pressure waves and convert them into output signals. During the transmission process of sound pressure waves, they are affected by the surrounding environment and are prone to generate echoes, which affects the communication quality.

[0003] Existing echo cancellation methods generally use a linear adaptive filter to model the acoustic propagation path of the room, then use the model and a reference signal to estimate the echo signal, and finally subtract the estimated echo signal from the microphone received signal to achieve the effect of echo cancellation.

[0004] In practical applications, a reference microphone is usually placed next to the speaker to obtain the nonlinear characteristics of the speaker for compensation. However, the sound waves emitted by the speaker can easily cause vibration of the structural parts in the communication loop, affecting the accuracy of the received signal. Summary of the invention

[0005] The embodiments of the present application provide an echo cancellation method, apparatus and communication device, which can improve the quality of audio transmission. The technical solution is as follows:

[0006] In a first aspect, an embodiment of the present application provides an echo cancellation method, comprising the following steps:

[0007] Broadcast remote signals using loudspeakers;

[0008] Acquire a first signal from a receiving microphone, a second signal from a reference microphone, and a third signal from a vibration sensor; wherein the reference microphone is disposed adjacent to the speaker, and the vibration sensor is disposed adjacent to the receiving microphone;

[0009] Based on the far-end signal and the second signal, construct a nonlinear model of the speaker and obtain nonlinear parameters of the nonlinear model;

[0010] Multiplying the nonlinear parameter and the far-end signal to obtain a nonlinear signal;

[0011] splicing the nonlinear signal and the third signal to obtain a dual-channel reference signal;

[0012] Based on the dual-channel reference signal and the first signal, acquiring a propagation path of a signal;

[0013] Convolving the dual-channel reference signal and the equivalent propagation path to obtain an echo interference signal;

[0014] The echo interference signal is eliminated from the first signal to obtain an echo cancellation signal.

[0015] Optionally, the step of obtaining the echo interference signal includes:

[0016] The dual-channel reference signal and the equivalent propagation path are convolved to obtain an echo interference signal in the following manner:

[0017]

[0018] Among them, echo I (n) represents the echo interference signal at time n, X I (nk) represents the nkth dual-channel reference signal, represents the kth equivalent propagation path, and K represents the number of k.

[0019] Optionally, the step of constructing a nonlinear model of the loudspeaker based on the far-end signal and the second signal includes:

[0020] The nonlinear model of the loudspeaker is constructed as follows:

[0021]

[0022] Where h is the echo path coefficient between the loudspeaker and the reference microphone, a i represents the model coefficient of the i-th order, and N is the order of the linear filter;

[0023] The step of obtaining the nonlinear parameters of the nonlinear model comprises:

[0024] Construct N linear filters and obtain the echo path coefficients h1,...,h N ;

[0025] Substitute the echo path coefficient into the nonlinear model of the loudspeaker to obtain the nonlinear parameters of the nonlinear model in the following manner:

[0026]

[0027] Where i = 1, ..., N, ρ i represents the i-th order nonlinear parameter.

[0028] Optionally, the step of acquiring the nonlinear signal includes:

[0029] To obtain a nonlinear signal, proceed as follows:

[0030]

[0031] Among them, XN represents a nonlinear signal, ρ i represents the i-th order nonlinear parameter, X represents the far-end signal, and N represents the order of the linear filter.

[0032] Optionally, the speaker is connected to a power amplifier;

[0033] Before the step of broadcasting a remote signal by using a loudspeaker, the step further includes:

[0034] A remote signal is received and output to the power amplifier.

[0035] Optionally, the step of splicing the nonlinear signal and the third signal to obtain a dual-channel reference signal includes:

[0036] Feature vectors are extracted from the nonlinear signal and the third signal, and vector concatenation is performed to obtain a dual-channel reference signal.

[0037] Optionally, the step of acquiring the propagation path of the signal includes:

[0038] Obtain the estimated value of the propagation path at time n in the following way:

[0039]

[0040] Among them, X I (nk) represents the dual-channel reference signal at time n, represents the estimated value of the propagation path at time n, represents the estimated value of the propagation path at time n-1, μ represents the coefficient, 0<μ<2; e1(n) represents the estimated value of the echo cancellation signal at time n, which is obtained as follows:

[0041]

[0042] Wherein, K represents the number of k, and y(n) represents the first signal at time n.

[0043] Optionally, the step of eliminating the echo interference signal from the first signal to obtain an echo cancellation signal includes:

[0044] The echo interference signal is subtracted from the first signal to obtain an echo cancellation signal.

[0045] In a second aspect, an embodiment of the present application provides an echo cancellation device, including:

[0046] A broadcast module, used to broadcast remote signals using a loudspeaker;

[0047] A signal acquisition module, used to acquire a first signal from a receiving microphone, a second signal from a reference microphone, and a third signal from a vibration sensor; wherein the reference microphone is disposed adjacent to the speaker, and the vibration sensor is disposed adjacent to the receiving microphone;

[0048] A nonlinear parameter acquisition module, configured to construct a nonlinear model of the speaker and acquire nonlinear parameters of the nonlinear model based on the remote signal and the second signal;

[0049] A nonlinear signal acquisition module, used for multiplying the nonlinear parameter and the remote signal to acquire a nonlinear signal;

[0050] A dual-channel reference signal acquisition module, used for splicing the nonlinear signal and the third signal to obtain a dual-channel reference signal;

[0051] An equivalent propagation path acquisition module, used to acquire a signal propagation path based on the dual-channel reference signal and the first signal;

[0052] An echo interference signal acquisition module, used to convolve the dual-channel reference signal and the equivalent propagation path to obtain an echo interference signal;

[0053] The echo cancellation module is used to eliminate the echo interference signal from the first signal to obtain an echo cancellation signal.

[0054] In a third aspect, an embodiment of the present application provides a communication device, comprising: a speaker, a reference microphone, a receiving microphone, a vibration sensor, a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the reference microphone is arranged adjacent to the speaker, the vibration sensor is arranged adjacent to the receiving microphone, and when the processor executes the computer program, the steps of the echo cancellation method as described in any one of the above items are implemented.

[0055] In an embodiment of the present application, a second signal is received by using a reference microphone arranged adjacent to a speaker, and a third signal is received by using a vibration sensor adjacent to a receiving microphone. A nonlinear signal for reflecting the nonlinear characteristics of the speaker is obtained according to the remote signal and the second signal. A dual-channel reference signal combining the nonlinear characteristics of the speaker and the device vibration signal caused by the speaker is obtained by splicing the nonlinear signal and the third signal. A propagation path of the signal is obtained based on the dual-channel reference signal and the first signal. The dual-channel reference signal and the equivalent propagation path are convolved to obtain an echo interference signal. The echo interference signal can effectively reflect the echo interference received during the transmission of the remote signal. By eliminating the echo interference signal from the first signal, the problem of incomplete echo elimination caused by the reception distortion caused by the speaker distortion and the vibration of the receiving microphone is avoided. The echo can be effectively eliminated to improve the audio transmission quality.

[0056] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is an application scenario diagram of an echo cancellation method in one embodiment of the present invention;

[0058] Figure 2 A flowchart of an echo cancellation method in one embodiment of the present invention;

[0059] Figure 3 It is a schematic diagram of the structure of an echo cancellation device in one embodiment of the present invention;

[0060] Figure 4 The figure is a schematic diagram of the structure of a communication device in one embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the objectives, technical solutions and advantages of the present application more clear, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0062] It should be clear that the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present application.

[0063] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms of "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0064] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0065] In addition, in the description of this application, unless otherwise specified, "several" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0066] The embodiments of the present application can be applied in Figure 1 In the communication device shown, the communication device includes a speaker 1, a reference microphone 2, a receiving microphone 3, a vibration sensor 4 and a power amplifier 5. The far-end signal is amplified by the power amplifier 5 and then transmitted to the receiving microphone 3 through the speaker 1. The reference microphone 2 is arranged adjacent to the speaker 1 for sensing the far-end signal broadcast from the speaker 1; the receiving microphone 3 is arranged at a farther distance from the speaker 1 than the reference microphone 2; the vibration sensor 4 is arranged adjacent to the receiving microphone 3 for acquiring a third signal of the receiving microphone 3.

[0067] The loudspeaker 1 may be replaced by any instrument or device that can convert electrical signals into sound energy.

[0068] The vibration sensor 4 may be a piezoelectric sensor or other instrument or device that can convert a vibration signal into an electrical signal.

[0069] In one embodiment, "close to" refers to a distance less than 5 mm. In other embodiments, "close to" may also be other distances that can meet the characteristics of the audio signal to be picked up by the embodiments of the present application.

[0070] As shown in the figure, the far-end signal, the second signal received by the reference microphone and the third signal received by the vibration sensor are used as reference signals for adjusting the output of the first signal. According to the far-end signal, the second signal and the third signal, the echo interference in the first signal is eliminated and an echo cancellation signal is output to improve the audio transmission quality.

[0071] like Figure 2 As shown, the embodiment of the present application provides an echo cancellation method, comprising the following steps:

[0072] Step S1: broadcasting a remote signal using a loudspeaker;

[0073] The far-end signal is the audio signal that the loudspeaker wants to send to the far-end, and the loudspeaker receives and broadcasts the far-end signal.

[0074] Step S2: Acquire a first signal from a receiving microphone, a second signal from a reference microphone, and a third signal from a vibration sensor;

[0075] The first signal is an audio signal received by the receiving microphone, including the broadcast signal of the speaker and background noise, and the second signal is an audio signal received by the reference microphone.

[0076] The receiving microphone receives the sound wave signal from the loudspeaker to generate vibration, and the vibration sensor is used to detect the vibration signal of the receiving microphone.

[0077] Step S3: constructing a nonlinear model of the speaker based on the far-end signal and the second signal and acquiring nonlinear parameters of the nonlinear model;

[0078] The nonlinear characteristics of a loudspeaker refer to the nonlinear vibration of the loudspeaker under the action of a large signal, which causes harmonic distortion and modulation distortion in the playback signal of the loudspeaker, affecting the audio quality.

[0079] The nonlinear parameters can be used to determine the nonlinear vibration of the speaker. In one embodiment, the nonlinear parameters of the speaker can be obtained by constructing a nonlinear model based on the far-end signal and the second signal. Specifically, the step of constructing the nonlinear model of the speaker based on the far-end signal and the second signal includes:

[0080] The nonlinear model of the loudspeaker is constructed as follows:

[0081]

[0082] Where h is the echo path coefficient between the loudspeaker and the reference microphone, a i represents the model coefficient of the i-th order, and N is the order of the linear filter;

[0083] The step of obtaining the nonlinear parameters of the nonlinear model comprises:

[0084] Construct N linear filters and obtain the echo path coefficients h1,...,h N ;

[0085] Substitute the echo path coefficient into the nonlinear model of the loudspeaker to obtain the nonlinear parameters of the nonlinear model in the following manner:

[0086]

[0087] Where i = 1, ..., N, ρ i represents the i-th order nonlinear parameter.

[0088] Step S4: multiplying the nonlinear parameter and the far-end signal to obtain a nonlinear signal;

[0089] The nonlinear signal is the echo signal generated during signal transmission due to the nonlinear characteristics of the loudspeaker, and can be obtained using the nonlinear parameters of the loudspeaker.

[0090] By eliminating the nonlinear signal from the first signal, the transmission quality of the output signal can be effectively improved. In one embodiment, the step of obtaining the nonlinear signal includes:

[0091] To obtain a nonlinear signal, proceed as follows:

[0092]

[0093] Among them, X N represents a nonlinear signal, ρ i represents the i-th order nonlinear parameter, X represents the far-end signal, and N represents the order of the linear filter.

[0094] Step S5: splicing the nonlinear signal and the third signal to obtain a dual-channel reference signal;

[0095] The dual-channel reference signal refers to a reference signal formed by combining an echo signal for describing the nonlinear characteristics of a loudspeaker and an echo signal of a receiving microphone caused by the loudspeaker.

[0096] In one embodiment, the nonlinear signal and the third signal are concatenated by superimposing the nonlinear signal and the third signal. In another embodiment, the step of constructing a dual-channel reference signal according to the nonlinear signal and the third signal includes extracting feature vectors from the nonlinear signal and the third signal and concatenating the feature vectors to obtain a dual-channel reference signal.

[0097] Step S6: acquiring a signal propagation path based on the dual-channel reference signal and the first signal;

[0098] The propagation path of the signal can be obtained by using a linear adaptive filter to derive the propagation path based on the first signal and the dual-channel signal. In one embodiment, the step of obtaining the propagation path of the signal includes:

[0099] Obtain the estimated value of the propagation path at time n in the following way:

[0100]

[0101] Among them, X I (nk) represents the dual-channel reference signal at time n, represents the estimated value of the propagation path at time n, represents the estimated value of the propagation path at time n-1, μ represents the coefficient, 0<μ<2; e1(n) represents the estimated value of the echo cancellation signal at time n, which is obtained as follows:

[0102]

[0103] Wherein, K represents the number of k, and y(n) represents the first signal at time n.

[0104] Step S7: performing convolution on the dual-channel reference signal and the equivalent propagation path to obtain an echo interference signal;

[0105] By convolving the dual-channel reference signal and the propagation path of the propagation process, the echo interference signal in the first signal can be obtained. Specifically, the echo interference signal is obtained in the following manner:

[0106]

[0107] Among them, echo I (n) represents the echo interference signal at time n, X I (nk) represents the nkth dual-channel reference signal, represents the kth equivalent propagation path, and K represents the number of k.

[0108] Step S8: Eliminate the echo interference signal from the first signal to obtain an echo cancellation signal

[0109] In one embodiment, the echo interference signal may be inverted and superimposed with the first signal to eliminate the echo interference signal in the first signal. In another embodiment, the echo cancellation signal may be obtained by subtracting the echo interference signal from the first signal.

[0110] The echo cancellation signal can be transmitted to a computer, tablet computer, mobile phone, interactive smart tablet or other instrument or device that can convert electrical signals into sound energy for audio playback or voice calls, thereby improving communication quality.

[0111] In an embodiment of the present application, a second signal is received by using a reference microphone arranged adjacent to a speaker, and a third signal is received by using a vibration sensor adjacent to a receiving microphone. A nonlinear signal for reflecting the nonlinear characteristics of the speaker is obtained according to the remote signal and the second signal. A dual-channel reference signal combining the nonlinear characteristics of the speaker and the device vibration signal caused by the speaker is obtained by splicing the nonlinear signal and the third signal. A propagation path of the signal is obtained based on the dual-channel reference signal and the first signal. The dual-channel reference signal and the equivalent propagation path are convolved to obtain an echo interference signal. The echo interference signal can effectively reflect the echo interference received during the transmission of the remote signal. By eliminating the echo interference signal from the first signal, the problem of incomplete echo elimination caused by the reception distortion caused by the speaker distortion and the vibration of the receiving microphone is avoided. The echo can be effectively eliminated to improve the audio transmission quality.

[0112] In one embodiment, before the step of broadcasting the remote-end signal by using a loudspeaker, the method further includes: receiving the remote-end signal and outputting the remote-end signal to a power amplifier.

[0113] The power amplifier is used to amplify the remote signal and output the amplified remote signal to the speaker to improve the power amplification effect of the speaker.

[0114] like Figure 3 As shown, the embodiment of the present application also provides an echo cancellation device, including:

[0115] Broadcast module 1, used for broadcasting remote signals using a loudspeaker;

[0116] A signal acquisition module 2, used to acquire a first signal from a receiving microphone, a second signal from a reference microphone, and a third signal from a vibration sensor; wherein the reference microphone is disposed adjacent to the speaker, and the vibration sensor is disposed adjacent to the receiving microphone;

[0117] A nonlinear parameter acquisition module 3, configured to construct a nonlinear model of the loudspeaker and acquire nonlinear parameters of the nonlinear model based on the remote signal and the second signal;

[0118] A nonlinear signal acquisition module 4, configured to multiply the nonlinear parameter and the remote signal to acquire a nonlinear signal;

[0119] A dual-channel reference signal acquisition module 5, used for splicing the nonlinear signal and the third signal to obtain a dual-channel reference signal;

[0120] An equivalent propagation path acquisition module 6, configured to acquire a signal propagation path based on the dual-channel reference signal and the first signal;

[0121] An echo interference signal acquisition module 7 is used to convolve the dual-channel reference signal and the equivalent propagation path to obtain an echo interference signal;

[0122] The echo cancellation module 8 is used to eliminate the echo interference signal from the first signal to obtain an echo cancellation signal.

[0123] It should be noted that the echo cancellation device provided in the above embodiment only uses the division of the above functional modules as an example when executing the echo cancellation method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the echo cancellation device and the echo cancellation method provided in the above embodiment belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0124] like Figure 4 As shown, an embodiment of the present application also provides a communication device, including: a speaker 100, a reference microphone 200, a receiving microphone 300, a vibration sensor 400, a memory 500, a processor 600, and a computer program stored in the memory 500 and executable by the processor 600, the reference microphone 200 is arranged adjacent to the speaker 100, the vibration sensor 400 is arranged adjacent to the receiving microphone 300, and the processor 600 implements the steps of the echo cancellation method as described in any one of the above when executing the computer program.

[0125] The communication device may be a computing device such as a computer, a tablet computer, a mobile phone, an interactive smart tablet, a laptop computer or a server, and the echo cancellation method is integrated into the communication device as one of its functions.

[0126] The speaker 100 may also be replaced by any instrument or device that can convert electrical signals into sound energy.

[0127] The vibration sensor 400 may be a piezoelectric sensor or other instrument or device that can convert a vibration signal into an electrical signal.

[0128] In one embodiment, the communication device further includes a power amplifier, which is used to amplify the remote signal and output the amplified remote signal to the speaker to improve the power amplification effect of the speaker.

[0129] The present invention is not limited to the above-mentioned embodiments. If various changes or modifications to the present invention do not depart from the spirit and scope of the present invention, and if these changes and modifications fall within the scope of the claims and equivalent technologies of the present invention, the present invention is also intended to include these changes and modifications.

Claims

1. An echo cancellation method, characterized in that: The following steps are involved: Broadcast remote signals using loudspeakers; Acquire a first signal from a receiving microphone, a second signal from a reference microphone, and a third signal from a vibration sensor; wherein the reference microphone is disposed adjacent to the speaker, and the vibration sensor is disposed adjacent to the receiving microphone; constructing a nonlinear model for reflecting the nonlinear characteristics of the speaker based on the far-end signal and the second signal, and acquiring nonlinear parameters of the nonlinear model according to the nonlinear model; Multiplying the nonlinear parameter and the far-end signal to obtain a nonlinear signal; Extracting feature vectors from the nonlinear signal and the third signal and performing vector concatenation to obtain a dual-channel reference signal; Based on the dual-channel reference signal and the first signal, a linear adaptive filter is used to derive a propagation path to obtain a propagation path of the signal; Convolving the dual-channel reference signal and the propagation path to obtain an echo interference signal; Eliminating the echo interference signal from the first signal to obtain an echo-cancelled signal; The step of constructing a nonlinear model of the loudspeaker based on the far-end signal and the second signal comprises: The nonlinear model of the loudspeaker is constructed as follows: Where X1(n) represents the second signal at time n, X represents the remote signal, and h i (k) represents the kth element of the echo path coefficient between the ith loudspeaker and the reference microphone, and h i (k) = a i h(k),a i represents the i-th order model coefficient, h(k) represents the k-th element of the echo path coefficient h, N is the order of the linear filter, and K represents the number of k; The step of obtaining the nonlinear parameters of the nonlinear model comprises: Construct N linear filters and obtain N echo path coefficients h1,...,h N ; Substitute the echo path coefficient into the nonlinear model of the loudspeaker to obtain the nonlinear parameters of the nonlinear model in the following manner: Where i = 1, ..., N, ρ i represents the i-th order nonlinear parameter.

2. The echo cancellation method according to claim 1, characterized in that: The step of obtaining the echo interference signal comprises: The dual-channel reference signal and the propagation path are convolved to obtain an echo interference signal in the following manner: Among them, echo I (n) represents the echo interference signal at time n, X I (nk) represents the nkth dual-channel reference signal, represents the kth propagation path, and K represents the number of k.

3. The echo cancellation method according to claim 1, characterized in that: The step of obtaining the nonlinear signal comprises: To obtain a nonlinear signal, proceed as follows: Among them, X N represents a nonlinear signal, ρ i represents the i-th order nonlinear parameter, X represents the far-end signal, N represents the order of the linear filter, X i Represents the far-end signal X raised to the power of i.

4. The echo cancellation method according to claim 1, characterized in that: The speaker is connected to a power amplifier; Before the step of broadcasting a remote signal by using a loudspeaker, the step further includes: A remote signal is received and output to the power amplifier.

5. The echo cancellation method according to claim 1, characterized in that: The step of obtaining the propagation path of the signal comprises: Obtain the propagation path estimate at time n in the following manner: Among them, X I (nk) represents the dual-channel reference signal at time n, represents the estimated value of the propagation path at time n, represents the estimated value of the propagation path at time n-1, μ represents the coefficient, 0<μ<2; e(n) represents the estimated value of the echo cancellation signal at time n, which is obtained as follows: Wherein, K is the number of vectors k, and y(n) represents the first signal at time n.

6. The echo cancellation method according to claim 1, characterized in that: The step of eliminating the echo interference signal from the first signal to obtain an echo cancellation signal comprises: The echo interference signal is subtracted from the first signal to obtain an echo cancellation signal.

7. An echo cancellation device, characterized in that: include: A broadcast module, used to broadcast remote signals using a loudspeaker; A signal acquisition module, used to acquire a first signal from a receiving microphone, a second signal from a reference microphone, and a third signal from a vibration sensor; wherein the reference microphone is disposed adjacent to the speaker, and the vibration sensor is disposed adjacent to the receiving microphone; A nonlinear parameter acquisition module is used to construct a nonlinear model reflecting the nonlinear characteristics of the loudspeaker based on the remote signal and the second signal, and to obtain nonlinear parameters of the nonlinear model according to the nonlinear model: the nonlinear model of the loudspeaker is constructed in the following manner: Where X1(n) represents the second signal at time n, X represents the remote signal, and h i (k) represents the kth element of the echo path coefficient between the ith loudspeaker and the reference microphone, and h i (k) = a i h(k),a i represents the i-th order model coefficient, h(k) represents the k-th element of the echo path coefficient h, N is the order of the linear filter, and K represents the number of k; The step of obtaining the nonlinear parameters of the nonlinear model comprises: Construct N linear filters and obtain N echo path coefficients h1,...,h N ; Substitute the echo path coefficient into the nonlinear model of the loudspeaker to obtain the nonlinear parameters of the nonlinear model in the following manner: Where i = 1, ..., N, ρ i represents the nonlinear parameter of the i-th order; A nonlinear signal acquisition module, used for multiplying the nonlinear parameter and the remote signal to acquire a nonlinear signal; A dual-channel reference signal acquisition module, used for extracting characteristic vectors from the nonlinear signal and the third signal and performing vector concatenation to obtain a dual-channel reference signal; An equivalent propagation path acquisition module, used to derive a propagation path using a linear adaptive filter based on the dual-channel reference signal and the first signal to acquire a signal propagation path; An echo interference signal acquisition module, used to convolve the dual-channel reference signal and the propagation path to obtain an echo interference signal; The echo cancellation module is used to eliminate the echo interference signal from the first signal to obtain an echo cancellation signal.

8. A communication device, characterized in that: include: A speaker, a reference microphone, a receiving microphone, a vibration sensor, a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the reference microphone is arranged adjacent to the speaker, the vibration sensor is arranged adjacent to the receiving microphone, and when the processor executes the computer program, the steps of the echo cancellation method as described in any one of claims 1 to 6 are implemented.

Images