CN111508464A - Filtering parameter self-updating method, filter, equipment and storage medium - Google Patents

Filtering parameter self-updating method, filter, equipment and storage medium Download PDF

Info

Publication number
CN111508464A
CN111508464A CN202010290783.7A CN202010290783A CN111508464A CN 111508464 A CN111508464 A CN 111508464A CN 202010290783 A CN202010290783 A CN 202010290783A CN 111508464 A CN111508464 A CN 111508464A
Authority
CN
China
Prior art keywords
error
energy
filter
error signal
projection order
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010290783.7A
Other languages
Chinese (zh)
Other versions
CN111508464B (en
Inventor
王维
王广新
杨汉丹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Youjie Zhixin Technology Co ltd
Original Assignee
Shenzhen Youjie Zhixin Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Youjie Zhixin Technology Co ltd filed Critical Shenzhen Youjie Zhixin Technology Co ltd
Priority to CN202010290783.7A priority Critical patent/CN111508464B/en
Publication of CN111508464A publication Critical patent/CN111508464A/en
Application granted granted Critical
Publication of CN111508464B publication Critical patent/CN111508464B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
    • G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
    • G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
    • G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
    • G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
    • G10K11/1785—Methods, e.g. algorithms; Devices
    • G10K11/17853—Methods, e.g. algorithms; Devices of the filter
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
    • G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
    • G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
    • G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
    • G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
    • G10K11/1787—General system configurations
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
    • H04R1/1083—Reduction of ambient noise
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R3/00—Circuits for transducers, loudspeakers or microphones
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
    • G10K2210/00—Details of active noise control [ANC] covered by G10K11/178 but not provided for in any of its subgroups
    • G10K2210/10—Applications
    • G10K2210/108—Communication systems, e.g. where useful sound is kept and noise is cancelled
    • G10K2210/1081—Earphones, e.g. for telephones, ear protectors or headsets
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
    • G10K2210/00—Details of active noise control [ANC] covered by G10K11/178 but not provided for in any of its subgroups
    • G10K2210/30—Means
    • G10K2210/301—Computational
    • G10K2210/3028—Filtering, e.g. Kalman filters or special analogue or digital filters

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Acoustics & Sound (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Soundproofing, Sound Blocking, And Sound Damping (AREA)

Abstract

The invention provides a filtering parameter self-updating method, a filter, equipment and a storage medium, which are applied to an affine projection filter in an active noise reduction system with an error microphone, wherein the projection order of the filter is updated based on the energy of an offset error, and the offset error is the difference value between an error signal and system noise; updating the step length based on the energy of the offset error and the energy of the error signal; when the active noise reduction system starts to work, the estimated offset error energy is large, the projection order is updated to the maximum, the step length update is close to the maximum step length under the projection order, the convergence speed is improved, and the system stability is guaranteed; and when the estimated offset error energy is small, updating the projection order to the minimum, wherein the step length update is close to the step length minimum under the projection order, so that the steady-state error is smaller. The invention realizes dynamic adjustment of the projection order and has small calculation amount.

Description

Filtering parameter self-updating method, filter, equipment and storage medium
Technical Field
The present invention relates to the field of filter technologies, and in particular, to a method, a filter, a device, and a storage medium for self-updating a filter parameter.
Background
When the ANC (Active Noise Cancellation, Active Noise control) earphone structure design is not optimal, the primary channel impulse response and the secondary channel impulse response are very long, or the interference Noise frequency band is wide, the ANC earphone convergence speed is slow, and the Noise reduction amplitude is low.
In the existing ANC algorithm, the convergence speed of the FX L MS-based improved algorithm, such as FXN L MS, VSS-FXN L MS and other algorithms, is faster than that of FXN L MS, although the calculation amount is not increased much, the effect is improved to a limited extent, and the balance between the convergence speed and the steady-state error is not well designed, and for example, some FXR L S algorithms can provide faster convergence speed and smaller steady-state error, but the matrix calculation consumes large resources and is not suitable for the use of light-weight earphones and other devices, and some domain-transforming methods, such as frequency domain ANC, sub-band ANC and the like, have better effect, but increase the system complexity and are easy to introduce extra time delay.
Disclosure of Invention
The invention mainly aims to provide a filtering parameter self-updating method, a filter, equipment and a storage medium, aiming at overcoming the defects that the convergence speed is high, the steady-state error is small and the calculated amount is small which cannot be realized at the same time at present.
In order to achieve the above object, the present invention provides a filter with self-updated filter parameters, wherein the filter is an affine projection filter in an active noise reduction system with an error microphone, and the filter parameters are a projection order and a step length;
the filter parameters of the filter are updated by the following method:
Figure BDA0002450315510000021
wherein N is a projection order which is an integer, NmaxTo set maximum projection order, NminFor a set minimum projection order, β for controlling the decay rateThe parameters are set to be in a predetermined range,
Figure BDA0002450315510000022
estimating the offset error energy, wherein the offset error is the difference between an error signal acquired by an error microphone and system noise;
mu(n)＝ρ(n)*mumax(N)+(1-ρ(n))*mumin(N)；
Figure BDA0002450315510000023
where mu (n) is the step size, mumax(N)、mumin(N) represents a maximum step size and a minimum step size in the projection order N, respectively, where ρ (N) is a convergence factor,
Figure BDA0002450315510000024
to estimate the energy of the error signal.
Further, the estimated misadjustment error energy is calculated as follows:
Figure BDA0002450315510000025
Figure BDA0002450315510000026
Figure BDA0002450315510000027
where α is the smoothing coefficient, the regularization parameter,
Figure BDA0002450315510000028
energy of an input signal picked up by a reference microphone in an active noise reduction system, e (n) the error signal,
Figure BDA0002450315510000029
is an input vector of the input signal.
Further, the energy of the estimated error signal is calculated as follows:
Figure BDA00024503155100000210
the invention also provides a filtering parameter self-updating method, which is applied to an active noise reduction system with an error microphone, wherein the active noise reduction system comprises an affine projection filter, and the filtering parameters are projection orders and step lengths, and the method comprises the following steps:
acquiring an error signal acquired by the error microphone;
calculating to obtain an offset error according to the difference value of the error signal and the system noise;
calculating the energy of the misadjustment error to obtain estimated misadjustment error energy, and calculating the energy of the error signal to obtain the energy of the estimated error signal;
updating the projection order according to the estimated misadjustment error energy; updating the step length according to the estimated offset error energy and the energy of the estimated error signal;
wherein, the updating mode of the projection order is as follows:
Figure BDA0002450315510000031
wherein N is a projection order which is an integer, NmaxTo set maximum projection order, NminTo set minimum projection order, β to control decay rate parameter,
Figure BDA0002450315510000032
to estimate the offset error energy;
mu(n)＝ρ(n)*mumax(N)+(1-ρ(n))*mumin(N)；
Figure BDA0002450315510000033
where mu (n) is the step size, mumax(N)、mumin(N) respectively represent the projectionThe maximum step size and the minimum step size when the number of the shadow is N, rho (N) is a convergence factor,
Figure BDA0002450315510000034
to estimate the energy of the error signal.
Further, the active noise reduction system further includes a reference microphone, and before the step of calculating an offset error according to a difference between the error signal and system noise, the method further includes:
acquiring an input signal acquired by the reference microphone;
the calculating the energy of the misadjustment error to obtain the energy of the estimated misadjustment error comprises the following steps:
Figure BDA0002450315510000035
Figure BDA0002450315510000036
Figure BDA0002450315510000037
where α is the smoothing coefficient, the regularization parameter,
Figure BDA0002450315510000038
energy of an input signal picked up by a reference microphone in an active noise reduction system, e (n) the error signal,
Figure BDA0002450315510000039
is an input vector of the input signal.
Further, the calculating the energy of the error signal to obtain the energy of the estimation error signal includes:
Figure BDA00024503155100000310
further, the calculating an offset error according to the difference between the error signal and the system noise includes:
Figure BDA0002450315510000041
wherein c (n) is the offset error, e (n) is the error signal, v (n) is the system noise,
Figure BDA0002450315510000042
is an input vector of the input signal,
Figure BDA0002450315510000043
in order to be the theoretical optimal weight value,
Figure BDA0002450315510000044
is a weight vector.
Further, the weight vector updating method is as follows:
Figure BDA0002450315510000045
where mu (n) is the step size at time n, which is the regularization parameter, (. cndot.)HFor the conjugate transpose operation, a is a matrix of data blocks of size N × L, L is the filter length,
Figure BDA0002450315510000046
is an error vector of size N x 1.
The invention also provides a device comprising a memory in which a computer program is stored and a processor which, when executing the computer program, carries out the steps of the method according to any one of the preceding claims.
The invention also provides a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the method of any of the above.
The filtering parameter self-updating method, the filter, the equipment and the storage medium provided by the invention are applied to an affine projection filter in an active noise reduction system with an error microphone, wherein the projection order of the filter is updated based on the energy of an offset error, and the offset error is the difference value between an error signal and system noise; updating the step length based on the energy of the offset error and the energy of the error signal; when the active noise reduction system starts to work, the estimated offset error energy is large, the projection order is updated to the maximum, the step length update is close to the maximum step length under the projection order, the convergence speed is improved, and the system stability is guaranteed; and when the estimated offset error energy is small, updating the projection order to the minimum, wherein the step length update is close to the step length minimum under the projection order, so that the steady-state error is smaller. The invention realizes dynamic adjustment of the projection order and has small calculation amount.
Drawings
FIG. 1 is a schematic diagram illustrating an algorithm of an active noise reduction system according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating the steps of a method for self-updating filter parameters according to an embodiment of the present invention;
FIG. 3 is a block diagram of a filtering parameter self-updating apparatus according to an embodiment of the present invention;
fig. 4 is a block diagram illustrating the structure of an apparatus according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides a filter with self-updating filter parameters, wherein the filter is an affine projection filter in an active noise reduction system with an error microphone, and the filter parameters are a projection order and a step length;
the filter parameters of the filter are updated by the following method:
Figure BDA0002450315510000051
wherein N is a projection order which is an integer, NmaxTo set maximum projection order, NminTo set minimum projection order, β to control decay rate parameter,
Figure BDA0002450315510000052
estimating the offset error energy, wherein the offset error is the difference between an error signal acquired by an error microphone and system noise; the system noise is denoted as v (n).
In this embodiment, based on the above-mentioned update formula of the projection order, when the active noise reduction system starts to work, the offset error energy is estimated because the offset error is large
Figure BDA0002450315510000053
Very big, then
Figure BDA0002450315510000054
Very small, close to 0, N ≈ N in the above formulamaxAnd updating the projection order to enable the active noise reduction system to be rapidly converged.
When the convergence of the active noise reduction system is close to a stable state, the offset error is very small, and the corresponding estimated offset error
Figure BDA0002450315510000055
Is very small, then
Figure BDA0002450315510000056
Close to 1, N ≈ N in the above formulaminAnd updating the projection order to make the active noise reduction system self-adaptively reduce the order so as to obtain smaller steady-state error.
Therefore, in this embodiment, the projection order of the affine projection filter can be dynamically adjusted by using the above projection order updating method, and the affine projection filter can be quickly converged when the misadjustment error is large, and can adaptively reduce the order when the misadjustment error is reduced, so as to obtain a smaller steady-state error. In the above updating process, the amount of calculation is small.
The step updating method of the filter is as follows:
mu(n)＝ρ(n)*mumax(N)+(1-ρ(n))*mumin(N)；
Figure BDA0002450315510000061
where mu (n) is the step size, mumax(N)、mumin(N) represents a maximum step size and a minimum step size in the projection order N, respectively, where ρ (N) is a convergence factor,
Figure BDA0002450315510000062
to estimate the energy of the error signal. Wherein, mumaxThe value of (N) is less than the theoretical maximum step size under the respective conditions.
In this embodiment, in practical application, the noise of the active noise reduction system is very small compared to the ambient noise, so that
Figure BDA0002450315510000063
By definition, the offset error energy is estimated when the active noise reduction system starts to work
Figure BDA0002450315510000064
Close to the energy of the estimation error signal, the convergence factor ρ (n) is close to 1. At this time, the step length is close to the maximum value of the step length when the projection order is N, and the convergence speed is further improved;
when the active noise reduction system is close to convergence, the offset error is very small, and the offset error energy is estimated
Figure BDA0002450315510000065
Close to 0 and the output error is close to Ev2(n)]And if the convergence factor rho (n) is close to 0 at the moment, updating the step size at the moment to be the minimum step size under the current projection order, and further reducing the steady-state error.
In the embodiment, the convergence factor rho (n) is used as an auxiliary parameter, the size of the step length is dynamically adjusted, and the stability of the system is improved. And in the step length updating process, the calculated amount is small. In the embodiment, the projection order and the step size are automatically adjusted, so that the method has the characteristics of higher convergence rate and smaller steady-state error compared with the traditional method.
In one embodiment, the estimated misadjustment error energy is calculated as follows:
Figure BDA0002450315510000066
the correlation between the cross-correlation energy between the input signal and the error signal and the estimated offset error energy, the energy of the input signal, is expressed in this formula.
Figure BDA0002450315510000067
The equation is the cross-correlation between the input signal and the error signal.
Figure BDA0002450315510000068
The formula expresses the way the energy of the input signal is calculated.
Where α is the smoothing coefficient, the regularization parameter,
Figure BDA0002450315510000071
energy of an input signal picked up by a reference microphone in an active noise reduction system, e (n) the error signal,
Figure BDA0002450315510000072
is an input vector of the input signal. In the embodiment, the cross-correlation between the error signal and the input signal is used to effectively cancel the influence of the system noise v (n), so that the robustness of the system is stronger.
In an embodiment, the energy of the error signal is estimated by a first-order recursive smoothing method, and the energy of the estimated error signal is calculated as follows:
Figure BDA0002450315510000073
in one embodiment, the calculation process of the misalignment error is as follows:
Figure BDA0002450315510000074
in this embodiment, a cross-correlation of the offset error and the input signal is obtained, where c (n) is the offset error, e (n) is the error signal, v (n) is the system noise,
Figure BDA0002450315510000075
is an input vector of the input signal,
Figure BDA0002450315510000076
in order to be the theoretical optimal weight value,
Figure BDA0002450315510000077
is a weight vector.
In this embodiment, the weight vector updating method is as follows:
Figure BDA0002450315510000078
where mu (n) is the step size at time n, which is the regularization parameter, (. cndot.)HFor the conjugate transpose operation, a is a matrix of data blocks of size N × L, L is the filter length,
Figure BDA0002450315510000079
is an error vector of size N x 1.
Wherein A is represented as:
Figure BDA00024503155100000710
referring to fig. 1, a schematic diagram of the active noise reduction system in an embodiment is shown, and the active noise reduction system is based on a feedforward network structure, where the filter used is an affine projection filter. Specifically, the active noise reduction system comprises a reference microphone and an error microphoneTwo sensors, using reference microphone to obtain primary noise x (n) (input signal), error microphone to obtain error signal e (n), in the invention, transfer functions of primary channel P (z) and secondary channel S (z) are obtained in advance by off-line system identification, that is, transfer functions of FIG. 2
Figure BDA00024503155100000711
For the primary noise collected by the reference microphone, the primary noise is vector-expressed as an input vector, and is defined as follows:
Figure BDA0002450315510000081
wherein L is the filter length;
the input vector passes through the
Figure BDA0002450315510000082
The transfer of the transfer function of (a) is then obtained:
Figure BDA0002450315510000083
Figure BDA0002450315510000084
after the input vector is filtered by the primary channel p (z), the following results are obtained:
Figure BDA0002450315510000085
the reference signal d (n) is the sum of the primary noise filtered by the primary channel and the system noise, namely:
d(n)＝x2(n)+v(n)；
the output of the input signal filtered by the filter w (z) is y, which is expressed as:
Figure BDA0002450315510000086
after being filtered by a stimulation channel, y' is expressed as:
Figure BDA0002450315510000087
the error signal e (n) can be expressed as:
e(n)＝d(n)-y'(n)；
further, an error vector of size N1
Figure BDA0002450315510000088
Expressed as:
Figure BDA0002450315510000089
the reference signal vector is represented as:
Figure BDA00024503155100000810
therefore, the update mode of the weight vector is obtained:
Figure BDA00024503155100000811
further, the invention obtains the imbalance error based on the weight imbalance error tracking algorithm of the cross-correlation energy estimation;
further tracking error energy estimation, defining the estimated misadjustment error energy as:
Figure BDA00024503155100000812
where E is the mathematical expectation symbol.
Referring to fig. 2, an embodiment of the present invention further provides a filtering parameter self-updating method, which is applied to an active noise reduction system having an error microphone, where the error microphone is used to collect an error signal, and the system may further include a reference microphone used to collect primary noise (an input signal); the active noise reduction system comprises an affine projection filter, and the filter parameters are a projection order and a step length, and the method comprises the following steps:
step S1, acquiring an error signal acquired by the error microphone;
step S2, calculating to obtain an offset error according to the difference value of the error signal and the system noise;
step S3, calculating the energy of the misadjustment error to obtain the energy of the estimated misadjustment error, and calculating the energy of the error signal to obtain the energy of the estimated error signal;
step S4, updating the projection order according to the estimated misadjustment error energy; updating the step length according to the estimated offset error energy and the energy of the estimated error signal;
in this embodiment, the offset error is calculated and the offset error energy is estimated, so as to update the projection order according to the estimated offset error energy.
Wherein, the updating mode of the projection order is as follows:
Figure BDA0002450315510000091
wherein N is a projection order which is an integer, NmaxTo set maximum projection order, NminTo set minimum projection order, β to control decay rate parameter,
Figure BDA0002450315510000092
to estimate the offset error energy; the system noise is denoted as v (n).
In this embodiment, based on the above-mentioned update formula of the projection order, when the active noise reduction system starts to work, the offset error energy is estimated because the offset error is large
Figure BDA0002450315510000093
Very big, then
Figure BDA0002450315510000094
Very small, close to 0, N ≈ N in the above formulamaxAnd updating the projection order to enable the active noise reduction system to be rapidly converged.
When the convergence of the active noise reduction system is close to a stable state, the offset error is very small, and the corresponding estimated offset error
Figure BDA0002450315510000095
Is very small, then
Figure BDA0002450315510000096
Close to 1, N ≈ N in the above formulaminAnd updating the projection order to make the active noise reduction system self-adaptively reduce the order so as to obtain smaller steady-state error.
Therefore, in this embodiment, the projection order of the affine projection filter can be dynamically adjusted by using the above projection order updating method, and the affine projection filter can be quickly converged when the misadjustment error is large, and can adaptively reduce the order when the misadjustment error is reduced, so as to obtain a smaller steady-state error. In the above updating process, the amount of calculation is small.
In this embodiment, the step size is updated according to the estimated offset error energy and the estimated error signal energy. The specific updating mode of the step length is as follows:
mu(n)＝ρ(n)*mumax(N)+(1-ρ(n))*mumin(N)；
Figure BDA0002450315510000101
where mu (n) is the step size, mumax(N)、mumin(N) represents a maximum step size and a minimum step size in the projection order N, respectively, where ρ (N) is a convergence factor,
Figure BDA0002450315510000102
to estimate the energy of the error signal. Wherein, mumax(N) is less than the theoretical maximum step length under the respective conditions。
In this embodiment, in practical application, the noise of the active noise reduction system is very small compared to the ambient noise, so that
Figure BDA0002450315510000103
By definition, the offset error energy is estimated when the active noise reduction system starts to work
Figure BDA0002450315510000104
Close to the energy of the estimation error signal, the convergence factor ρ (n) is close to 1. At this time, the step length is close to the maximum value of the step length when the projection order is N, and the convergence speed is further improved;
when the active noise reduction system is close to convergence, the offset error is very small, and the offset error energy is estimated
Figure BDA0002450315510000105
Close to 0 and the output error is close to Ev2(n)]And if the convergence factor rho (n) is close to 0 at the moment, updating the step size at the moment to be the minimum step size under the current projection order, and further reducing the steady-state error.
In the embodiment, the convergence factor rho (n) is used as an auxiliary parameter, the size of the step length is dynamically adjusted, and the stability of the system is improved. And in the step length updating process, the calculated amount is small. In the embodiment, the projection order and the step size are automatically adjusted, so that the method has the characteristics of higher convergence rate and smaller steady-state error compared with the traditional method.
In an embodiment, the active noise reduction system further includes a reference microphone, and before the step S2 of calculating an offset error according to a difference between the error signal and system noise, the method further includes:
step S1a, acquiring an input signal acquired by the reference microphone;
the calculating the energy of the misadjustment error to obtain the energy of the estimated misadjustment error comprises the following steps:
Figure BDA0002450315510000106
Figure BDA0002450315510000111
Figure BDA0002450315510000112
where α is the smoothing coefficient, the regularization parameter,
Figure BDA0002450315510000113
energy of an input signal picked up by a reference microphone in an active noise reduction system, e (n) the error signal,
Figure BDA0002450315510000114
is an input vector of the input signal. In the embodiment, the cross-correlation between the error signal and the input signal is used to effectively cancel the influence of the system noise v (n), so that the robustness of the system is stronger.
In this embodiment, estimating the energy of the error signal by using a first-order recursive smoothing method, where the calculating the energy of the error signal to obtain the energy of the estimated error signal includes:
Figure BDA0002450315510000115
in this embodiment, the calculating the offset error according to the difference between the error signal and the system noise includes:
Figure BDA0002450315510000116
in this embodiment, a cross-correlation of the offset error and the input signal is obtained, where c (n) is the offset error, e (n) is the error signal, v (n) is the system noise,
Figure BDA0002450315510000117
is an input vector of the input signal,
Figure BDA0002450315510000118
in order to be the theoretical optimal weight value,
Figure BDA0002450315510000119
is a weight vector.
Further, the weight vector updating method is as follows:
Figure BDA00024503155100001110
where mu (n) is the step size at time n, which is the regularization parameter, (. cndot.)HFor the conjugate transpose operation, a is a matrix of data blocks of size N × L, L is the filter length,
Figure BDA00024503155100001111
is an error vector of size N x 1.
Wherein A is represented as:
Figure BDA00024503155100001112
for the explanation of the algorithm principle of the active noise reduction system, please refer to fig. 1 and the description in the above embodiment, which will not be repeated herein.
Referring to fig. 3, an embodiment of the present invention further provides a filtering parameter self-updating apparatus, which is applied to an active noise reduction system having an error microphone, where the error microphone is used to collect an error signal, and the system may further include a reference microphone used to collect primary noise (an input signal); the active noise reduction system comprises an affine projection filter, the filter parameters are a projection order and a step length, and the device comprises:
an obtaining unit 10, configured to obtain an error signal collected by the error microphone;
a first calculating unit 20, configured to calculate an offset error according to a difference between the error signal and system noise;
a second calculating unit 30, configured to calculate an energy of the misadjustment error to obtain an estimated misadjustment error energy, and calculate an energy of the error signal to obtain an energy of the estimated error signal;
an updating unit 40, configured to update the projection order according to the estimated misadjustment error energy; updating the step length according to the estimated offset error energy and the energy of the estimated error signal;
the manner of updating the projection order by the updating unit 40 is as follows:
Figure BDA0002450315510000121
wherein N is a projection order which is an integer, NmaxTo set maximum projection order, NminTo set minimum projection order, β to control decay rate parameter,
Figure BDA0002450315510000122
to estimate the offset error energy; the system noise is denoted as v (n).
The way in which the update unit 40 updates the step size is as follows:
mu(n)＝ρ(n)*mumax(N)+(1-ρ(n))*mumin(N)；
Figure BDA0002450315510000123
where mu (n) is the step size, mumax(N)、mumin(N) represents a maximum step size and a minimum step size in the projection order N, respectively, where ρ (N) is a convergence factor,
Figure BDA0002450315510000124
to estimate the energy of the error signal. Wherein, mumaxThe value of (N) is less than the theoretical maximum step size under the respective conditions.
In this embodiment, please refer to the above method embodiments for specific implementation of the above units, which will not be described herein again.
Referring to fig. 4, an embodiment of the present invention further provides a device, where the device may be an earphone, and an internal structure of the device may be as shown in fig. 4. The device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores a computer program and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the device is used to store signals, filter parameter self-updating programs, etc. The network interface of the device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of filter parameter self-updating.
It will be understood by those skilled in the art that the structure shown in fig. 4 is only a block diagram of a portion of the structure associated with the inventive arrangements, and does not constitute a limitation on the computer apparatus to which the inventive arrangements are applied.
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements a filtering parameter self-updating method. It is to be understood that the computer-readable storage medium in the present embodiment may be a volatile-readable storage medium or a non-volatile-readable storage medium.
In summary, the filtering parameter self-updating method, the filter, the device and the storage medium provided in the embodiments of the present invention are applied to an affine projection filter in an active noise reduction system with an error microphone, wherein a projection order of the filter is updated based on energy of an offset error, and the offset error is a difference between an error signal and system noise; updating the step length based on the energy of the offset error and the energy of the error signal; when the active noise reduction system starts to work, the estimated offset error energy is large, the projection order is updated to the maximum, the step length update is close to the maximum step length under the projection order, the convergence speed is improved, and the system stability is guaranteed; and when the estimated offset error energy is small, updating the projection order to the minimum, wherein the step length update is close to the step length minimum under the projection order, so that the steady-state error is smaller. The invention realizes dynamic adjustment of the projection order and has small calculation amount.
It will be understood by those of ordinary skill in the art that all or part of the processes of the methods of the embodiments described above may be implemented by a computer program that may be stored on a non-volatile computer readable storage medium, which when executed, may include the processes of the embodiments of the methods described above, wherein any reference to memory, storage, database or other medium provided and used in the embodiments of the present invention may include non-volatile and/or volatile memory.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above description is only for the preferred embodiment of the present invention and is not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (10)

1. A filter with self-updated filter parameters is characterized in that the filter is an affine projection filter in an active noise reduction system with an error microphone, and the filter parameters are a projection order and a step length;
the filter parameters of the filter are updated by the following method:
Figure FDA0002450315500000011
wherein N is a projection order which is an integer, NmaxTo set maximum projection order, NminTo set minimum projection order, β to control decay rate parameter,
Figure FDA0002450315500000012
estimating the offset error energy, wherein the offset error is the difference between an error signal acquired by an error microphone and system noise;
mu(n)＝ρ(n)*mumax(N)+(1-ρ(n))*mumin(N)；
Figure FDA0002450315500000013
where mu (n) is the step size, mumax(N)、mumin(N) represents a maximum step size and a minimum step size in the projection order N, respectively, where ρ (N) is a convergence factor,
Figure FDA0002450315500000014
to estimate the energy of the error signal.
2. The filter parameter self-updating filter of claim 1, wherein the estimated misadjustment error energy is calculated as follows:
Figure FDA0002450315500000015
Figure FDA0002450315500000016
Figure FDA0002450315500000017
where α is the smoothing coefficient, the regularization parameter,
Figure FDA0002450315500000018
energy of an input signal picked up by a reference microphone in an active noise reduction system, e (n) the error signal,
Figure FDA0002450315500000019
is an input vector of the input signal.
3. The filter parameter self-updating filter of claim 2, wherein the energy of the estimated error signal is calculated as follows:
Figure FDA0002450315500000021
4. a filtering parameter self-updating method is applied to an active noise reduction system with an error microphone, the active noise reduction system comprises an affine projection filter, the filtering parameters are a projection order and a step size, and the method comprises the following steps:
acquiring an error signal acquired by the error microphone;
calculating to obtain an offset error according to the difference value of the error signal and the system noise;
calculating the energy of the misadjustment error to obtain estimated misadjustment error energy, and calculating the energy of the error signal to obtain the energy of the estimated error signal;
updating the projection order according to the estimated misadjustment error energy; updating the step length according to the estimated offset error energy and the energy of the estimated error signal;
wherein, the updating mode of the projection order is as follows:
Figure FDA0002450315500000022
wherein N is a projection order which is an integer, NmaxTo set maximum projection order, NminTo set minimum projection order, β to control decay rate parameter,
Figure FDA0002450315500000023
to estimate the offset error energy;
mu(n)＝ρ(n)*mumax(N)+(1-ρ(n))*mumin(N)；
Figure FDA0002450315500000024
where mu (n) is the step size, mumax(N)、mumin(N) represents a maximum step size and a minimum step size in the projection order N, respectively, where ρ (N) is a convergence factor,
Figure FDA0002450315500000025
to estimate the energy of the error signal.
5. The method of claim 4, wherein the active noise reduction system further comprises a reference microphone, and the step of calculating the offset error according to the difference between the error signal and the system noise further comprises:
acquiring an input signal acquired by the reference microphone;
the calculating the energy of the misadjustment error to obtain the energy of the estimated misadjustment error comprises the following steps:
Figure FDA0002450315500000031
Figure FDA0002450315500000032
Figure FDA0002450315500000033
where α is the smoothing coefficient, the regularization parameter,
Figure FDA0002450315500000034
energy of an input signal picked up by a reference microphone in an active noise reduction system, e (n) the error signal,
Figure FDA0002450315500000035
is an input vector of the input signal.
6. The filter parameter self-updating method of claim 5, wherein said calculating the energy of the error signal to obtain the energy of the estimated error signal comprises:
Figure FDA0002450315500000036
7. the method of claim 4, wherein calculating an offset error according to a difference between the error signal and a system noise comprises:
Figure FDA0002450315500000037
wherein c (n) is the offset error, e (n) is the error signal, v (n) is the system noise,
Figure FDA0002450315500000038
is an input vector of the input signal,
Figure FDA0002450315500000039
in order to be the theoretical optimal weight value,
Figure FDA00024503155000000310
is a weight vector.
8. The method of claim 7, wherein the weight vector is updated in a manner of:
Figure FDA00024503155000000311
where mu (n) is the step size at time n, which is the regularization parameter, (. cndot.)HFor the conjugate transpose operation, a is a matrix of data blocks of size N × L, L is the filter length,
Figure FDA00024503155000000312
is an error vector of size N x 1.
9. An apparatus comprising a memory and a processor, the memory having stored therein a computer program, wherein the processor, when executing the computer program, implements the steps of the method of any of claims 4 to 8.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 4 to 8.
CN202010290783.7A 2020-04-14 2020-04-14 Filtering parameter self-updating method, filter, device and storage medium Active CN111508464B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010290783.7A CN111508464B (en) 2020-04-14 2020-04-14 Filtering parameter self-updating method, filter, device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010290783.7A CN111508464B (en) 2020-04-14 2020-04-14 Filtering parameter self-updating method, filter, device and storage medium

Publications (2)

Publication Number Publication Date
CN111508464A true CN111508464A (en) 2020-08-07
CN111508464B CN111508464B (en) 2023-07-14

Family

ID=71877527

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010290783.7A Active CN111508464B (en) 2020-04-14 2020-04-14 Filtering parameter self-updating method, filter, device and storage medium

Country Status (1)

Country Link
CN (1) CN111508464B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112013285A (en) * 2020-08-26 2020-12-01 辽宁石油化工大学 Method and device for detecting pipeline leakage point, storage medium and terminal

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6137881A (en) * 1997-02-28 2000-10-24 Texas Instruments Incorporated Adaptive filtering method and apparatus employing modified fast affine projection algorithm
US20130322638A1 (en) * 2012-06-02 2013-12-05 Yuan Ze University Method for acoustic echo cancellation and system thereof
KR101432426B1 (en) * 2013-02-14 2014-08-22 한양대학교 산학협력단 Method and apparatus for determining filter coefficients for an adaptive filter
CN106448695A (en) * 2016-09-28 2017-02-22 天津大学 Double-end call robust variable-order variable-step-size affine projection method
MX2017012678A (en) * 2017-10-02 2018-05-11 Inst Politecnico Nacional Compact and configurable method and device for performing multichannel adaptive filtering based on related projections with encoded error and evolutionary order.
CN108630216A (en) * 2018-02-15 2018-10-09 湖北工业大学 A kind of MPNLMS acoustic feedback suppressing methods based on dual microphone model
CN109089004A (en) * 2018-08-14 2018-12-25 西南交通大学 A kind of collection person's adaptive echo cancellation method based on joint entropy induction

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6137881A (en) * 1997-02-28 2000-10-24 Texas Instruments Incorporated Adaptive filtering method and apparatus employing modified fast affine projection algorithm
US20130322638A1 (en) * 2012-06-02 2013-12-05 Yuan Ze University Method for acoustic echo cancellation and system thereof
KR101432426B1 (en) * 2013-02-14 2014-08-22 한양대학교 산학협력단 Method and apparatus for determining filter coefficients for an adaptive filter
CN106448695A (en) * 2016-09-28 2017-02-22 天津大学 Double-end call robust variable-order variable-step-size affine projection method
MX2017012678A (en) * 2017-10-02 2018-05-11 Inst Politecnico Nacional Compact and configurable method and device for performing multichannel adaptive filtering based on related projections with encoded error and evolutionary order.
CN108630216A (en) * 2018-02-15 2018-10-09 湖北工业大学 A kind of MPNLMS acoustic feedback suppressing methods based on dual microphone model
CN109089004A (en) * 2018-08-14 2018-12-25 西南交通大学 A kind of collection person's adaptive echo cancellation method based on joint entropy induction

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
MEDEL C H 等: "The Galileo Ground Segment Integrity Algorithms: Design and Performance", 《INTERNATIONAL JOURNAL OF NAVIGATION & OBSERVATION》 *
SHIN H C 等: "Variable step-size NLMS and affine projection algorithms", 《IEEE SIGNAL PROCESSING LETTERS》 *
孟庆萍 等: "基于可变阶数的集员滤波仿射投影算法[", 《无线电工程》 *
师黎明: "仿射投影类自适应滤波算法的改进算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112013285A (en) * 2020-08-26 2020-12-01 辽宁石油化工大学 Method and device for detecting pipeline leakage point, storage medium and terminal
CN112013285B (en) * 2020-08-26 2022-04-19 辽宁石油化工大学 Method and device for detecting pipeline leakage point, storage medium and terminal

Also Published As

Publication number Publication date
CN111508464B (en) 2023-07-14

Similar Documents

Publication Publication Date Title
Thenua et al. Simulation and performance analysis of adaptive filter in noise cancellation
JP2018527625A (en) Secondary path phase estimation in active noise control
CN110402540B (en) Active noise reduction method, device, chip, active control system and storage medium
EP3437090A1 (en) Adaptive modeling of secondary path in an active noise control system
CN111048106B (en) Pickup method and apparatus based on double microphones and computer device
JP2018527624A (en) Estimation of secondary path size in active noise control
CN103229237A (en) Signal processing device, signal processing method, and signal processing program
CN112037752A (en) Household appliance noise reduction method and device, computer equipment and storage medium
CN103716013A (en) Variable parameter proportion self-adaptive filter
JP2018518123A (en) Hearing aid system operating method and hearing aid system
Resende et al. A variable step-size NLMS algorithm with adaptive coefficient vector reusing
CN111508464A (en) Filtering parameter self-updating method, filter, equipment and storage medium
CN100524464C (en) Adaptive filter method and apparatus for improving speech quality of mobile communication apparatus
Bernardi et al. An all-frequency-domain adaptive filter with PEM-based decorrelation for acoustic feedback control
CN113132848A (en) Filter design method and device and in-ear active noise reduction earphone
Nishiyama A unified view of adaptive algorithms for finite impulse response filters using the H∞ framework
Antoñanzas et al. Affine-projection-like algorithm for active noise control over distributed networks
Munjal et al. RLS algorithm for acoustic echo cancellation
Bertrand et al. Distributed LCMV beamforming in wireless sensor networks with node-specific desired signals
CN107483029B (en) Method and device for adjusting length of adaptive filter in voip communication
Qu et al. An efficient L 0 norm constraint memory improved proportionate affine projection algorithm
CN113035222B (en) Voice noise reduction method and device, filter determination method and voice interaction equipment
Caiyun Study of speech enhancement system using combinational adaptive filtering
Ferrer et al. Fast exact variable order affine projection algorithm
Zheng et al. Combination of bias-compensated normalized LMS algorithm for system identification with noisy input

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant