CN113539228A

CN113539228A - Noise reduction parameter determination method and device, active noise reduction method and device

Info

Publication number: CN113539228A
Application number: CN202110873136.3A
Authority: CN
Inventors: 刘益帆; 徐银海
Original assignee: Beijing Ansheng Haolang Technology Co ltd
Current assignee: Beijing Ansheng Haolang Technology Co ltd
Priority date: 2021-07-30
Filing date: 2021-07-30
Publication date: 2021-10-22
Anticipated expiration: 2041-07-30
Also published as: CN113539228B

Abstract

The application provides a method for determining noise reduction parameters, which comprises the following steps: acquiring a noise reduction signal sample output by a loudspeaker and an error signal sample acquired by an error microphone; determining an error signal trend parameter sample corresponding to the error signal sample based on the error signal sample; respectively carrying out fuzzy quantization processing on the error signal sample, the error signal trend parameter sample and the noise reduction signal sample to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector; determining a fuzzy control mapping relation among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector; and determining the noise reduction parameters based on the fuzzy control mapping relation among the fuzzy vectors of the error signal samples, the fuzzy vectors of the error signal trend parameter samples, the fuzzy vectors of the noise reduction signal samples, so that the design cost of the system can be reduced, the active noise reduction effect can be improved, and the probability of overshoot can be effectively reduced.

Description

Noise reduction parameter determination method and device, active noise reduction method and device

Technical Field

The application relates to the technical field of active noise reduction, in particular to a noise reduction parameter determination method and device, and an active noise reduction method and device.

Background

In recent years, products having an active noise reduction function have been attracting attention. Noise reduction parameters of an existing Active Noise reduction (ANC) controller are designed based on an accurate solution of an acoustic path, that is, the acoustic path is regarded as a "system", and a system response is determined according to a relationship between an input acoustic signal and an output acoustic signal.

However, the objective calculation of the acoustic path is complicated, and the acoustic path may vary with the usage state of the product, which may easily result in inaccurate calculation of the acoustic path. Inaccurate calculation of the acoustic path seriously affects the noise reduction effect of the target area and even can not reduce the noise. In addition, overshoot phenomenon is easy to occur in the active noise reduction process.

Disclosure of Invention

In view of this, embodiments of the present application provide a noise reduction parameter determination method and apparatus, and an active noise reduction method and apparatus, which can reduce system design cost, improve active noise reduction effect of a product, and effectively reduce probability of overshoot occurring in an active noise reduction process.

According to a first aspect of an embodiment of the present application, there is provided a method for determining a noise reduction parameter, including: acquiring a noise reduction signal sample output by a loudspeaker and an error signal sample acquired by an error microphone; determining an error signal trend parameter sample corresponding to the error signal sample based on the error signal sample, wherein the error signal trend parameter sample is used for representing the subsequent operation trend of the error signal sample; respectively carrying out fuzzy quantization processing on the error signal sample, the error signal trend parameter sample and the noise reduction signal sample to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector; determining a fuzzy control mapping relation among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector; and determining the noise reduction parameters based on the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector, the noise reduction signal sample fuzzy vector and the fuzzy control mapping relation.

In an embodiment of the present application, the performing fuzzy quantization processing on the error signal sample, the error signal trend parameter sample, and the noise reduction signal sample to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector, and a noise reduction signal sample fuzzy vector respectively includes: respectively determining error signal fuzzy partition information and error signal trend parameter fuzzy partition information by taking the error signal sample and the error signal trend parameter sample as observed quantities of fuzzy control; based on the error signal fuzzy partition information, carrying out fuzzy quantization on the error signal sample into an error signal sample fuzzy vector; based on the fuzzy division information of the error signal trend parameter, fuzzily quantizing the error signal trend parameter sample into an error signal trend parameter sample fuzzy vector; determining fuzzy division information of the noise reduction signal by taking the noise reduction signal sample as a control quantity of fuzzy control; and based on the fuzzy partition information of the noise reduction signal, carrying out fuzzy quantization on the noise reduction signal samples into fuzzy vectors of the noise reduction signal samples.

In one embodiment of the present application, determining the error signal fuzzy partition information and the error signal trend parameter fuzzy partition information respectively by using the error signal samples and the error signal trend parameter samples as observed quantities of fuzzy control, includes: respectively determining an error signal quantization grade and an error signal discourse domain, and an error signal trend parameter quantization grade and an error signal trend parameter discourse domain by taking the error signal sample and the error signal trend parameter sample as observed quantities of fuzzy control; determining the membership degree of the error signal based on the quantization grade of the error signal and the discourse domain of the error signal, wherein the membership degree of the error signal is used for representing the degree of the accurate value of the discourse domain of the error signal which is subordinated to the quantization grade of the error signal; and determining the membership degree of the trend parameter of the error signal based on the quantization grade of the trend parameter of the error signal and the trend parameter domain of the error signal, wherein the membership degree of the trend parameter of the error signal is used for representing the degree of the accurate value of the trend parameter domain of the error signal which is subordinate to the quantization grade of the trend parameter of the error signal. And/or, determining the fuzzy division information of the noise reduction signal by taking the noise reduction signal sample as a control quantity of fuzzy control, wherein the fuzzy division information comprises the following steps: determining the quantization level of the noise reduction signal and the discourse domain of the noise reduction signal by taking the noise reduction signal sample as the control quantity of fuzzy control; and determining the membership degree of the noise reduction signal based on the quantization level of the noise reduction signal and the domain of the noise reduction signal, wherein the membership degree of the noise reduction signal is used for representing the degree of the accurate value of the domain of the noise reduction signal which is subordinate to the quantization level of the noise reduction signal.

In one embodiment of the present application, the fuzzy control mapping is used to eliminate the amplitude of the error signal samples while suppressing amplitude overshoot, and finally to keep the amplitude in a state of minimum convergence.

In one embodiment of the present application, determining a fuzzy control mapping relationship between the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector, and the noise reduction signal sample fuzzy vector comprises: if the quantization level of the error signal corresponding to the fuzzy vector of the error signal sample is negative large or negative small, and the quantization level of the trend parameter of the error signal corresponding to the fuzzy vector of the trend parameter sample of the error signal is negative large or negative small, the quantization level of the noise reduction signal corresponding to the fuzzy vector of the noise reduction signal sample is positive large; if the quantization level of the error signal corresponding to the fuzzy vector of the error signal sample is negative zero or positive zero, and the quantization level of the trend parameter of the error signal corresponding to the fuzzy vector of the trend parameter sample of the error signal is negative large or negative small, the quantization level of the noise reduction signal corresponding to the fuzzy vector of the noise reduction signal sample is positive small; and if the error signal quantization level corresponding to the error signal sample fuzzy vector is negative zero or positive zero, and the error signal trend parameter quantization level corresponding to the error signal trend parameter sample fuzzy vector is zero, the noise reduction signal quantization level corresponding to the noise reduction signal sample fuzzy vector is zero.

In one embodiment of the present application, determining a noise reduction parameter based on the error signal sample blur vector, the error signal trend parameter sample blur vector, the noise reduction signal sample blur vector, and the fuzzy control mapping relationship includes: aiming at each fuzzy control mapping relation in a plurality of fuzzy control mapping relations, carrying out Cartesian product operation on an error signal sample fuzzy vector and a noise reduction signal sample fuzzy vector which are mapped with each other and correspond to the fuzzy control mapping relation to obtain a first intermediate value matrix corresponding to the fuzzy control mapping relation, carrying out Cartesian product operation on an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector which are mapped with each other and correspond to the fuzzy control mapping relation to obtain a second intermediate value matrix corresponding to the fuzzy control mapping relation, and carrying out minimum operation on elements in the first intermediate value matrix and the second intermediate value matrix to obtain a third intermediate value matrix corresponding to the fuzzy control mapping relation; and carrying out maximum operation on elements of the third intermediate value matrix corresponding to the fuzzy control mapping relations respectively to obtain the noise reduction parameters.

According to a second aspect of the embodiments of the present application, there is provided an active noise reduction method, including: determining a current error signal trend parameter corresponding to the current error signal based on the current error signal acquired by the error microphone, wherein the current error signal trend parameter is used for representing the subsequent operation trend of the current error signal; fuzzy quantization processing is carried out on the current error signal and the current error signal trend parameter to obtain a current error signal fuzzy vector and a current error signal trend parameter fuzzy vector; determining a current noise reduction signal fuzzy vector corresponding to a current noise reduction signal based on the current error signal fuzzy vector, the current error signal trend parameter fuzzy vector and the noise reduction parameter, wherein the noise reduction parameter is determined based on the noise reduction parameter determination method of the first aspect; and carrying out deblurring quantization processing on the fuzzy vector of the current noise reduction signal to obtain the current noise reduction signal so that the speaker can play the current noise reduction signal.

In an embodiment of the present application, the performing deblurring quantization processing on a current error signal blur vector to obtain a current noise reduction signal includes: determining a current noise reduction signal discourse domain to which a current noise reduction signal fuzzy vector is subordinate by searching the maximum membership degree; and determining the current noise reduction signal corresponding to the current noise reduction signal domain based on the corresponding relation between the noise reduction signal domain and the noise reduction signal.

According to a third aspect of embodiments of the present application, there is provided a noise reduction parameter determination apparatus, including: the acquisition module is configured to acquire a noise reduction signal sample output by the loudspeaker and an error signal sample acquired by the error microphone; the first determination module is configured to determine an error signal trend parameter sample corresponding to the error signal sample based on the error signal sample, wherein the error signal trend parameter sample is used for representing a subsequent running trend of the error signal sample; the second determining module is configured to perform fuzzy quantization processing on the error signal sample, the error signal trend parameter sample and the noise reduction signal sample respectively to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector; a third determining module configured to determine a fuzzy control mapping relationship among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector; and the fourth determination module is configured to determine the noise reduction parameters based on the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector, the noise reduction signal sample fuzzy vector and the fuzzy control mapping relation.

According to a fourth aspect of embodiments of the present application, there is provided an active noise reduction device, including: the fifth determining module is configured to determine a current error signal trend parameter corresponding to the current error signal based on the current error signal acquired by the error microphone, wherein the current error signal trend parameter is used for representing a subsequent operation trend of the current error signal; the fuzzy quantization module is configured to perform fuzzy quantization processing on the current error signal and the current error signal trend parameter to obtain a current error signal fuzzy vector and a current error signal trend parameter fuzzy vector; a fuzzy inference module configured to determine a current noise reduction signal fuzzy vector corresponding to a current noise reduction signal based on the current error signal fuzzy vector, the current error signal trend parameter fuzzy vector and a noise reduction parameter, wherein the noise reduction parameter is determined based on the noise reduction parameter determination method according to the first aspect; and the de-blurring quantization module is configured to perform de-blurring quantization processing on the current noise reduction signal fuzzy vector to obtain a current noise reduction signal so that the speaker can play the current noise reduction signal.

The noise reduction parameter determining method provided by the embodiment of the application determines an error signal trend parameter sample for representing the subsequent running trend of the error signal sample based on the error signal sample, performs fuzzy quantization processing on the error signal sample, the noise reduction signal sample and the error signal trend parameter sample respectively to obtain an error signal sample fuzzy vector, a noise reduction signal sample fuzzy vector and an error signal trend parameter sample fuzzy vector, determines an error signal sample fuzzy vector, and determining the noise reduction parameters based on the fuzzy control mapping relation among the fuzzy vector of the error signal trend parameter sample, the fuzzy vector of the noise reduction signal sample and the fuzzy control mapping relation, so that the noise reduction parameters can be determined without calculating an acoustic path. The design cost of the active noise reduction system can be obviously reduced without establishing an accurate mathematical model of an acoustic path, and the active noise reduction process combined with the fuzzy control has the intelligence and the high efficiency of processing a nonlinear time-varying process based on the characteristics of the fuzzy control, so that the active noise reduction effect can be obviously improved. In addition, the follow-up running trend of the error signal sample is considered when the noise reduction parameters are determined, so that the probability of overshoot in the active noise reduction process can be effectively reduced.

Drawings

Fig. 1 is a block diagram of an active noise reduction system in the prior art that determines noise reduction parameters by establishing an acoustic path.

Fig. 2 is a schematic flow chart of a noise reduction parameter determining method according to an embodiment of the present application.

Fig. 3 is a schematic flow chart of a noise reduction parameter determining method according to an embodiment of the present application.

Fig. 4 is a schematic flow chart of a noise reduction parameter determining method according to an embodiment of the present application.

Fig. 5 is a schematic flow chart of a noise reduction parameter determining method according to an embodiment of the present application.

Fig. 6 is a schematic flow chart illustrating a method for determining a noise reduction parameter according to an embodiment of the present application.

Fig. 7 is a schematic flow chart illustrating a method for determining a noise reduction parameter according to an embodiment of the present application.

Fig. 8 is a schematic flowchart of an active noise reduction method according to an embodiment of the present application.

Fig. 9 is a block diagram of an active noise reduction system corresponding to the active noise reduction method provided in fig. 8.

Fig. 10 is a schematic structural diagram of a noise reduction parameter determination apparatus according to an embodiment of the present application.

Fig. 11 is a schematic structural diagram of a noise reduction parameter determination apparatus according to an embodiment of the present application.

Fig. 12 is a schematic structural diagram of an active noise reduction device according to an embodiment of the present application.

Fig. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The error microphone 120 is disposed near the target noise reduction region and configured to collect an error signal e, where the error signal e is a noise signal obtained in the target noise reduction region after noise reduction, the calculating module 130 is configured to determine a noise reduction signal y according to the error signal e, and the speaker 110 is configured to output the noise reduction signal y to reduce noise of the noise signal.

The space between the original noise signal x to the error microphone 120 forms a primary path, and the speaker 110 itself and the space between the speaker 110 to the error microphone 120 together constitute a secondary path. The primary path and the secondary path have respective transfer functions, where the transfer function of the primary path is a transfer function of the original noise signal x to the space between the error microphone 120, denoted as P, and the transfer function of the secondary path is a transfer function of the electro-acoustic conversion of the loudspeaker 110 and a transfer function of the diaphragm face of the loudspeaker 110 to the space between the error microphone 120, denoted as G. And determining a noise reduction parameter W by calculating G to establish a feedback active noise reduction system.

However, the objective calculation of the acoustic path is complicated, and the acoustic path may vary with the usage state of the product, which may easily result in inaccurate calculation of the acoustic path. Inaccurate calculation of the acoustic path seriously affects the noise reduction effect of the target area and even can not reduce the noise. In addition, overshoot phenomenon is easy to occur in the active noise reduction process. Therefore, the noise reduction parameter determination method provided by the embodiment of the application can determine the noise reduction parameters without calculating the acoustic path, and can effectively reduce the probability of overshoot in the active noise reduction process.

Specifically, an error signal trend parameter sample used for representing the subsequent running trend of the error signal sample is determined based on the error signal sample, fuzzy quantization processing is respectively carried out on the error signal sample, a noise reduction signal sample and the error signal trend parameter sample to obtain an error signal sample fuzzy vector, a noise reduction signal sample fuzzy vector and an error signal trend parameter sample fuzzy vector, a fuzzy control mapping relation among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector is determined, a noise reduction parameter is determined based on the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector, the noise reduction signal sample fuzzy vector and the fuzzy control mapping relation, and the noise reduction parameter can be determined without calculating an acoustic path. The design cost of the active noise reduction system can be obviously reduced without establishing an accurate mathematical model of an acoustic path, and the active noise reduction process combined with the fuzzy control has the intelligence and the high efficiency of processing a nonlinear time-varying process based on the characteristics of the fuzzy control, so that the active noise reduction effect can be obviously improved. In addition, the follow-up running trend of the error signal sample is considered when the noise reduction parameters are determined, so that the probability of overshoot in the active noise reduction process can be effectively reduced.

The noise reduction parameter determination method, the active noise reduction method, the noise reduction parameter determination device, the active noise reduction device, and the electronic device mentioned in the present application are further illustrated in conjunction with fig. 2 to 13.

Exemplary noise reduction parameter determination method

Fig. 2 is a schematic flow chart of a noise reduction parameter determining method according to an embodiment of the present application. As shown in fig. 2, the noise reduction parameter determination method includes the following steps.

Step 201: and acquiring a noise reduction signal sample output by the loudspeaker and an error signal sample acquired by the error microphone.

Specifically, the noise reduction signal samples output by the speaker are transferred to a target noise reduction region for noise reduction, and the noise signals within the target noise reduction region after noise reduction are collected by an error microphone (i.e., error signal samples), so that the noise reduction signal samples correspond to the error signal samples.

Step 202: and determining error signal trend parameter samples corresponding to the error signal samples based on the error signal samples.

In one embodiment, the error signal trend parameter samples are used to characterize a subsequent operating trend of the error signal samples.

Specifically, overshoot phenomenon easily occurs in the active noise reduction process, and the overshoot phenomenon is adjustment exceeding the adjustment range that should be adjusted for the error signal, for example: the sound pressure value of the current error signal is m Pa, and the purpose of active noise reduction is to suppress the sound pressure to the absolute value (2 x 10) of the lowest audible limit of human ears^-5Pa) but noise reduction is performed according to the noise reduction signal, the sound pressure value of the error signal may be reduced by n Pa (n is greater than m), and an overshoot phenomenon occurs. The overshoot phenomenon occurs because only the error signal sample (absolute sound pressure value) is considered, and the subsequent operation trend (relative sound pressure value) of the error signal sample is not considered. Therefore, the error signal trend parameter sample which can represent the subsequent running trend of the error signal sample is introduced in the process of determining the noise reduction parameters, so that the error signal sample and the error signal trend parameter sample are taken into consideration in the process of determining the noise reduction parameters, and the probability of overshoot phenomenon in the active noise reduction process is reduced.

It should be noted that the error trend parameter samples include error sample changes (i.e., the first derivative of the sound pressure values of the error signal samples). When the error signal trend parameter sample is an error signal sample change, the method for determining the error signal trend parameter sample based on the error signal sample is to calculate a 1 st derivative of the sound pressure value of the error signal sample.

Step 203: and respectively carrying out fuzzy quantization processing on the error signal sample, the error signal trend parameter sample and the noise reduction signal sample to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector.

In particular, fuzzy control is a control method based on fuzzy set theory, fuzzy linguistic variables and fuzzy logic reasoning, which simulates the human fuzzy reasoning and decision making process. In the process of fuzzy control, the experience of operators or experts is compiled into a fuzzy control rule, then signals are subjected to fuzzy quantization in real time, the signals subjected to fuzzy quantization are used as the input of the fuzzy control rule to complete fuzzy reasoning, and the output quantity obtained after the fuzzy reasoning is added to an actuator.

In the active noise reduction process combined with fuzzy control, an error signal sample, an error signal trend parameter sample and a noise reduction signal sample need to be converted into a fuzzy linguistic variable, so that fuzzy logic reasoning can be carried out. Therefore, the error signal samples, the error signal trend parameter samples, and the noise reduction signal sample blur need to be quantized into an error signal sample blur vector, an error signal trend parameter sample blur vector, and a noise reduction signal sample blur vector.

Step 204: and determining a fuzzy control mapping relation among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector.

Specifically, for performing fuzzy control, a fuzzy control rule needs to be established by depending on experience of an operator or an expert, and the fuzzy control rule is the core for performing fuzzy reasoning and decision-making subsequently. Based on the fuzzy control mapping relation, namely the fuzzy control rule, among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector is determined by combining the characteristics and experience of feedback type active noise reduction in the active noise reduction process of the fuzzy control.

It should be noted that the establishment principle of the fuzzy control mapping relationship is as follows: the fuzzy control rule comprises a quick response rule for dividing large error signal samples, a damping rule for preventing the error signal from being overshot and a steady-state rule for maintaining the error signal samples to be converged.

Step 205: and determining the noise reduction parameters based on the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector, the noise reduction signal sample fuzzy vector and the fuzzy control mapping relation.

Specifically, the process of determining the noise reduction parameters is a process of simulating fuzzy logic reasoning and decision, and fuzzy linguistic variables and fuzzy control rules are needed for the fuzzy reasoning and decision. Therefore, the noise reduction parameters are determined by synthesizing the error signal sample blur vector, the error signal trend parameter sample blur vector, the noise reduction signal sample blur vector, and the blur control mapping relation.

In the embodiment of the application, an error signal trend parameter sample used for representing the subsequent running trend of the error signal sample is determined based on the error signal sample, fuzzy quantization processing is respectively carried out on the error signal sample, a noise reduction signal sample and the error signal trend parameter sample to obtain an error signal sample fuzzy vector, a noise reduction signal sample fuzzy vector and an error signal trend parameter sample fuzzy vector, a fuzzy control mapping relation among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector is determined, a noise reduction parameter is determined based on the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector, the noise reduction signal sample fuzzy vector and the fuzzy control mapping relation, and the noise reduction parameter can be determined without calculating an acoustic path. The design cost of the active noise reduction system can be obviously reduced without establishing an accurate mathematical model of an acoustic path, and the active noise reduction process combined with the fuzzy control has the intelligence and the high efficiency of processing a nonlinear time-varying process based on the characteristics of the fuzzy control, so that the active noise reduction effect can be obviously improved. In addition, the follow-up running trend of the error signal sample is considered when the noise reduction parameters are determined, so that the probability of overshoot in the active noise reduction process can be effectively reduced.

Fig. 3 is a schematic flow chart of a noise reduction parameter determining method according to an embodiment of the present application. As shown in fig. 3, the step of performing fuzzy quantization processing on the error signal sample, the error signal trend parameter sample and the noise reduction signal sample to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector includes the following steps.

Step 301: and respectively determining error signal fuzzy partition information and error signal trend parameter fuzzy partition information by taking the error signal sample and the error signal trend parameter sample as observed quantities of fuzzy control.

Specifically, the feedback active noise reduction system adjusts the noise reduction signal sample output by the speaker through the error signal sample of the target noise reduction area acquired by the error microphone, so as to realize active noise reduction, therefore, the error signal sample is required to be taken as an observed quantity for fuzzy control, and meanwhile, in order to reduce the probability of overshoot, the subsequent operation trend of the error signal sample is also required to be considered, so that the error signal trend parameter sample is also required to be taken as an observed quantity for fuzzy control, and then the error signal sample and the error signal trend parameter sample are jointly taken as the observed quantity for fuzzy control, that is, the fuzzy controller is a bivariate fuzzy controller in the active noise reduction process. The error signal fuzzy partition information is information for fuzzy quantization of error signal samples, which is formulated by combining the characteristics of active noise reduction and experience, and is usually shown in the form of an error signal fuzzy partition table. The fuzzy division information of the error signal trend parameters is information which is formulated by combining the characteristics of active noise reduction and experience and is used for carrying out fuzzy quantization on the samples of the error signal trend parameters, and is usually displayed in the form of a fuzzy division table of the error signal trend parameters.

Step 302: the error signal sample blur is quantized into an error signal sample blur vector based on the error signal blur partition information.

Specifically, the error signal sample blur is quantized into an error signal sample blur vector by looking up in the error signal blur partition information.

Step 303: based on the fuzzy partition information of the error signal trend parameter, the fuzzy quantization of the error signal trend parameter sample is carried out to obtain a fuzzy vector of the error signal trend parameter sample.

Specifically, the error signal trend parameter sample blur is quantized into an error signal trend parameter sample blur vector by looking up in the error signal trend parameter blur partition information.

Step 304: and determining the fuzzy division information of the noise reduction signal by taking the noise reduction signal sample as the control quantity of fuzzy control.

Specifically, the feedback active noise reduction system realizes active noise reduction through noise reduction signal samples output by the loudspeaker. Therefore, the blur control is a control amount of the noise reduction signal sample. The noise reduction signal fuzzy partition information is information for performing fuzzy quantization on noise reduction signal samples, which is formulated by combining the characteristics of active noise reduction and experience, and is usually displayed in the form of a noise reduction signal fuzzy partition table.

Step 305: and based on the fuzzy partition information of the noise reduction signal, carrying out fuzzy quantization on the noise reduction signal samples into fuzzy vectors of the noise reduction signal samples.

Specifically, the noise reduction signal sample blur is quantized into a noise reduction signal sample blur vector by looking up in the noise reduction signal blur partition information.

In the embodiment of the application, error signal fuzzy partition information, error signal trend parameter fuzzy partition information and noise reduction signal fuzzy partition information are worked out by combining the characteristics and experience of active noise reduction, the error signal fuzzy partition information is inquired, an error signal sample is fuzzily quantized into an error signal sample fuzzy vector, the error signal trend parameter fuzzy partition information is inquired, the error signal trend parameter sample is fuzzily quantized into an error signal trend parameter sample fuzzy vector, the noise reduction signal fuzzy partition information is inquired, the noise reduction signal sample is fuzzily quantized into a noise reduction signal sample fuzzy vector, and therefore the error signal sample, the error signal trend parameter sample and the noise reduction signal sample are converted into fuzzy linguistic variables to carry out fuzzy logic reasoning.

Fig. 4 is a schematic flow chart of a noise reduction parameter determining method according to an embodiment of the present application. As shown in fig. 4, the step of determining the error signal fuzzy partition information and the error signal trend parameter fuzzy partition information respectively by taking the error signal samples and the error signal trend parameter samples as the observed quantities of the fuzzy control comprises the following steps.

Step 401: and respectively determining an error signal quantization grade and an error signal discourse domain, and an error signal trend parameter quantization grade and an error signal trend parameter discourse domain by taking the error signal sample and the error signal trend parameter sample as observed quantities of fuzzy control.

Specifically, the error signal quantization level is a level in which the error signal is described in a fuzzy language. The quantization level of the error signal trend parameter is used for describing the size of the error signal trend parameter by fuzzy language, namely, the size of the first derivative of the sound pressure value of the error signal is described by the fuzzy language. The quantization level is usually selected from large, medium and small, positive and negative zero, etc. to form fuzzy language. The error signal discourse domain is a set, elements in the set are discretization numerical values determined based on the magnitude of the physical quantity of the error signal, and each discretization numerical value corresponds to an accurate error signal. The argument of the error signal trend parameter is a set, and the elements in the set are discretization values determined based on the magnitude of the physical quantity of the error signal change, that is, the elements in the set are discretization values determined based on the magnitude of the first derivative value of the sound pressure value of the error signal, and each discretization value corresponds to the first derivative value of the sound pressure value of the error signal. Domains of discourse are commonly referred to by Arabic numerals and "plus and minus".

For example, the error signal quantization level and the error signal trend parameter quantization level are { negative, zero, positive }, { negative large, negative small, zero, positive small, positive large }, { negative large, negative medium, negative small, zero, positive small, positive medium, positive large } or { negative large, negative medium, negative small, negative zero, positive small, positive medium, positive large } or the like, abbreviated with english prefix { N, O, P }, { NB, NS, O, PS }, { NB, NM, NS, O, PS, PM, PB } or { NB, NM, NS, NO, PO, PS, PM, PB }, or the like. The error signal universe is { -2, -1, 0, +1, +2}, { -3, -2, -1, 0, +1, +2, +3} or { -4, -3, -2, -1, 0, +1, +2, +3, +4} etc.

Step 402: and determining the membership degree of the error signal based on the quantization grade of the error signal and the discourse domain of the error signal, wherein the membership degree of the error signal is used for representing the degree of the accurate value of the discourse domain of the error signal which is subordinated to the quantization grade of the error signal.

Specifically, in order to implement fuzzy quantization, a relationship is established between an error signal domain composed of the above precise values and the quantization levels of the error signals described by the fuzzy language, that is, the degree of membership of the precise values in the error signal domain to the respective quantization levels of the error signals is determined, and the degree of membership of the error signals is used for representing the degree of membership of the precise values of the error signal domain to the quantization levels of the error signals.

Step 403: and determining the membership degree of the trend parameter of the error signal based on the quantization grade of the trend parameter of the error signal and the trend parameter domain of the error signal, wherein the membership degree of the trend parameter of the error signal is used for representing the degree of the accurate value of the trend parameter domain of the error signal which is subordinate to the quantization grade of the trend parameter of the error signal.

Specifically, in order to implement fuzzy quantization, a relationship is established between an error signal trend parameter domain composed of the above precise values and an error signal trend parameter quantization level described by a fuzzy language, that is, the membership of the precise values in the error signal trend parameter domain to the respective error signal trend parameter quantization levels is determined, and the error signal trend parameter membership is used for representing the degree to which the precise values of the error signal trend parameter domain are subordinate to the error signal trend parameter quantization level.

For example, the error signal blur division information (error signal blur division table) is shown in table 1.

TABLE 1

The error signal tendency parameter blur division information (error signal tendency parameter blur division table) is shown in table 2.

TABLE 2

In the embodiment of the application, an error signal sample and an error signal trend parameter sample are taken as observed quantities of fuzzy control, an error signal quantization grade and an error signal discourse domain and an error signal trend parameter quantization grade and an error signal trend parameter discourse domain are determined, an error signal membership degree and an error signal trend parameter membership degree are determined, an error signal fuzzy partition table and an error signal trend parameter fuzzy partition table are formed, and the error signal sample and the error signal trend parameter sample are respectively subjected to fuzzy quantization into an error signal sample fuzzy vector and an error signal trend parameter sample fuzzy vector based on the error signal fuzzy partition table and the error signal trend parameter fuzzy partition table.

It should be noted that the division of the error signal trend parameter quantization level and the error signal quantization level may be the same or different, and the division of the error signal trend parameter domain and the error signal domain may be the same or different.

Fig. 5 is a schematic flow chart of a noise reduction parameter determining method according to an embodiment of the present application. As shown in fig. 5, the step of determining the fuzzy partition information of the noise reduction signal by taking the noise reduction signal sample as the control quantity of the fuzzy control comprises the following steps.

Step 501: and determining the quantization level of the noise reduction signal and the discourse domain of the noise reduction signal by taking the noise reduction signal sample as the control quantity of fuzzy control.

Step 502: and determining the membership degree of the noise reduction signal based on the quantization level of the noise reduction signal and the domain of the noise reduction signal, wherein the membership degree of the noise reduction signal is used for representing the degree of the accurate value of the domain of the noise reduction signal which is subordinate to the quantization level of the noise reduction signal.

Specifically, the method for determining the quantization level of the noise reduction signal, the discourse domain of the noise reduction signal and the membership degree of the noise reduction signal is the same as the method for determining the quantization level of the error signal, the discourse domain of the error signal and the membership degree of the error signal, and is not described herein again.

For example, the noise reduction signal blur division information (noise reduction signal blur division table) is shown in table 3.

TABLE 3

In the embodiment of the application, the noise reduction signal sample is used as the control quantity of fuzzy control, the quantization level of the noise reduction signal and the domain of the noise reduction signal are determined, the membership degree of the noise reduction signal is determined, a noise reduction signal fuzzy division table is formed, and based on the noise reduction signal fuzzy division table, the information of the noise reduction signal sample is subjected to fuzzy quantization into a noise reduction signal sample fuzzy vector.

The division of the noise reduction signal quantization level and the error signal quantization level may be the same or different, and the division of the noise reduction signal discourse domain and the error signal discourse domain may be the same or different.

Fig. 6 is a schematic flow chart illustrating a method for determining a noise reduction parameter according to an embodiment of the present application. As shown in fig. 6, the step of determining the fuzzy control mapping relationship among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector comprises the following steps.

Step 601: and if the quantization level of the error signal corresponding to the fuzzy vector of the error signal sample is negative large or negative small, and the quantization level of the error signal trend parameter corresponding to the fuzzy vector of the error signal trend parameter sample is negative large or negative small, the quantization level of the noise reduction signal corresponding to the fuzzy vector of the noise reduction signal sample is positive large.

Step 602: and if the quantization level of the error signal corresponding to the fuzzy vector of the error signal sample is negative zero or positive zero, and the quantization level of the trend parameter of the error signal corresponding to the fuzzy vector of the trend parameter sample of the error signal is negative large or negative small, the quantization level of the noise reduction signal corresponding to the fuzzy vector of the noise reduction signal sample is positive small.

Step 603: and if the error signal quantization level corresponding to the error signal sample fuzzy vector is negative zero or positive zero, and the error signal trend parameter quantization level corresponding to the error signal trend parameter sample fuzzy vector is zero, the noise reduction signal quantization level corresponding to the noise reduction signal sample fuzzy vector is zero.

In particular, the fuzzy control mapping relationship is typically composed of a set of fuzzy conditional statements of if-then structure. The error signal sample is represented by symbol e, and the error signal sample ambiguity vector is represented by symbol

The expression, the error signal trend parameter sample is expressed by symbol ec, and the error signal trend parameter sample fuzzy vector is expressed by symbol ec

Representing, the noise reduction signal sample by symbol y, the error signal sample fuzzy vector by symbol

The fuzzy control mapping relation is expressed by an if-then structure.

The fuzzy control mapping relationship is expressed as follows:

if...，then...

in the embodiment of the application, the fuzzy control mapping relation is expressed by a group of multiple conditional statements by combining the characteristics and experience of active noise reduction, so that noise reduction parameters obtained based on the error signal sample fuzzy vector and the error signal trend parameter sample fuzzy vector, the noise reduction signal sample fuzzy vector and the fuzzy control mapping relation are equivalent to noise reduction parameters obtained by calculating an acoustic path.

It should be noted that the fuzzy control mapping relationship may include not only the relationship represented by the above expression, but also relationships represented by other expressions.

Fig. 7 is a schematic flow chart illustrating a method for determining a noise reduction parameter according to an embodiment of the present application. As shown in fig. 7, the step of determining the noise reduction parameters based on the error signal sample blur vector, the error signal trend parameter sample blur vector, the noise reduction signal sample blur vector and the blur control mapping relation includes the following steps.

Step 701: aiming at each fuzzy control mapping relation in the plurality of fuzzy control mapping relations, carrying out Cartesian product operation on an error signal sample fuzzy vector and a noise reduction signal sample fuzzy vector which are mapped with each other and correspond to the fuzzy control mapping relation to obtain a first intermediate value matrix corresponding to the fuzzy control mapping relation, carrying out Cartesian product operation on an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector which are mapped with each other and correspond to the fuzzy control mapping relation to obtain a second intermediate value matrix corresponding to the fuzzy control mapping relation, and carrying out minimum operation on elements in the first intermediate value matrix and the second intermediate value matrix to obtain a third intermediate value matrix corresponding to the fuzzy control mapping relation.

Step 702: and carrying out maximum operation on elements of the third intermediate value matrix corresponding to the fuzzy control mapping relations respectively to obtain the noise reduction parameters.

In particular, noise reduction parameters

The method is a fuzzy relation matrix determined based on the fuzzy vector of the error signal sample, the fuzzy vector of the error signal trend parameter sample, the fuzzy vector of the noise reduction signal sample and the fuzzy control mapping relation, and the specific mathematical calculation method refers to the

steps

701 and 702.

For example, when the fuzzy control mapping relation is the expression from step 601 to step 603, the noise reduction parameter

The calculation formula of (2) is as follows:

wherein, "+" is the element to get the maximum operation; "x" represents the cartesian product of the vectors; "is the minimum operation taken by the element. "is a small operation taken by the elements in the ordered pair in the cartesian product.

In the embodiment of the application, based on the fuzzy control mapping relation, the fuzzy vector of the error signal sample, the fuzzy vector of the error signal trend parameter sample and the fuzzy vector of the noise reduction signal sample are calculated through the formula, so that the noise reduction parameter is obtained.

It should be noted that, when the quantization levels of the error signal samples, the error signal trend parameter samples, and the noise reduction signal samples and the division of the domain are fine enough, and the fuzzy control mapping relation is sufficient to meet the active noise reduction requirement, the noise reduction parameters obtained based on the fuzzy control

Linear filter obtained by infinite approximation of system characteristics through objective calculation

It should also be noted that the error trend parameter samples may include not only the error sample variation (i.e., the first derivative of the sound pressure value of the error sample), but also the variation of the error sample variation (i.e., the second derivative of the sound pressure value of the error sample). When the error signal trend parameter sample is the change of the error signal sample, the method for determining the error signal trend parameter sample based on the error signal sample is to calculate 1-order derivative of the sound pressure value of the error signal sample, and similarly, when the error signal trend parameter sample is the change of the error signal sample, the method for determining the error signal trend parameter sample based on the error signal sample is to calculate 2-order derivative of the sound pressure value of the error signal sample.

Fig. 8 is a schematic flowchart of an active noise reduction method according to an embodiment of the present application. Fig. 9 is a block diagram of an active noise reduction system corresponding to the active noise reduction method provided in fig. 8. As shown in fig. 8 and fig. 9, the active noise reduction method includes the following steps.

Step 801: and determining a current error signal trend parameter corresponding to the current error signal based on the current error signal acquired by the error microphone.

Specifically, in order to reduce the probability of occurrence of the overshoot phenomenon in the active noise reduction process, the subsequent operation trend of the current error signal sample needs to be considered, and therefore, a current error signal trend parameter corresponding to the current error signal needs to be obtained, and the first derivative of the sound pressure value of the current error signal is obtained in real time to obtain the current error signal trend parameter.

For example, the current error signal collected by the error microphone is e (n), and the derivative of the current error signal e (n) is obtained to obtain the current error signal trend parameter ec (n).

Step 802: and carrying out fuzzy quantization processing on the current error signal and the current error signal trend parameter to obtain a current error signal fuzzy vector and a current error signal trend parameter fuzzy vector.

Specifically, the current error signal and the current error signal trend parameter are both accurate quantities, and are required to be converted into a fuzzy language for fuzzy inference, so that the current error signal and the current error signal trend parameter are subjected to fuzzy quantization processing respectively based on preset error signal fuzzy partition information and preset error signal trend parameter fuzzy partition information to obtain a current error signal fuzzy vector and a current error signal trend parameter fuzzy vector. The specific process of obtaining the preset error signal fuzzy partition information and the preset error signal trend parameter fuzzy partition information in the noise reduction parameter determination method provided in any of the above embodiments is the same as that of obtaining the error signal fuzzy partition information and the error signal trend parameter fuzzy partition information, and details are not repeated here.

For example, the fuzzy quantization processing is performed on e (n), so as to obtain the fuzzy vector of the current error signal

Obtaining a fuzzy vector of the trend parameter of the current error signal by carrying out fuzzy quantization processing on ec (n)

Step 803: and determining a current noise reduction signal fuzzy vector corresponding to the current noise reduction signal based on the current error signal fuzzy vector, the current error signal trend parameter fuzzy vector and the noise reduction parameter.

Optionally, the noise reduction parameter is determined based on the noise reduction parameter determination method provided in any of the above embodiments.

Specifically, fuzzy reasoning is carried out on the current error signal fuzzy vector, the current error signal trend parameter fuzzy vector and the noise reduction parameter to obtain the current noise reduction signal fuzzy vector.

By way of example, to

And the noise reduction parameters obtained above

Fuzzy reasoning is carried out to obtain the fuzzy vector of the current noise reduction signal

Namely, it is

Step 804: and carrying out deblurring quantization processing on the fuzzy vector of the current noise reduction signal to obtain the current noise reduction signal so that the speaker can play the current noise reduction signal.

Specifically, the current noise reduction signal fuzzy vector obtained through fuzzy inference is a fuzzy quantity, and the current noise reduction signal fuzzy vector needs to be subjected to deblurring quantization processing to obtain an accurate physical signal, namely the current noise reduction signal.

In one embodiment, a current noise reduction signal discourse domain to which the current noise reduction signal fuzzy vector is subjected is determined by searching for the maximum degree of membership, and a current noise reduction signal corresponding to the current noise reduction signal discourse domain is determined based on the corresponding relationship between the noise reduction signal discourse domain and the noise reduction signal.

For example, the current noise reduction signal blur vector

Can be further expressed as:

and searching the maximum membership degree of 1, wherein the discourse domain of the current noise reduction signal is grade "-1", and obtaining the current noise reduction signal corresponding to grade "-1" according to the corresponding relation between the discourse domain of the noise reduction signal and the noise reduction signal. The correspondence between the noise reduction signal domain and the noise reduction signal is determined based on how to partition the domain when the noise reduction signal fuzzy partition information is obtained in the noise reduction parameter determination method provided in any of the above embodiments, and the specific process is not described herein again.

In the embodiment of the application, fuzzy quantization processing is carried out on the current error signal and the current error signal trend parameter to obtain a current error signal fuzzy vector and a current error signal trend parameter fuzzy vector, fuzzy reasoning is carried out on the current error signal fuzzy vector, the current error signal trend parameter fuzzy vector and the noise reduction parameter obtained in any one embodiment to obtain a current noise reduction signal fuzzy vector, de-fuzzy quantization processing is carried out on the current noise reduction signal fuzzy vector to obtain a current noise reduction signal, so that the loudspeaker plays the current noise reduction signal to reduce the noise of a target noise reduction area, and when the error microphone acquires an updated noise signal, the above processes are circulated until steady-state noise reduction is realized.

Exemplary noise reduction parameter determination apparatus

Fig. 10 is a schematic structural diagram of a noise reduction parameter determination apparatus according to an embodiment of the present application. As shown in fig. 10, the noise reduction parameter determination apparatus 100 includes: an obtaining module 101 configured to obtain a noise reduction signal sample output by a speaker and an error signal sample collected by an error microphone; the first determining module 102 is configured to determine an error signal trend parameter sample corresponding to the error signal sample based on the error signal sample, wherein the error signal trend parameter sample is used for characterizing a subsequent running trend of the error signal sample; the second determining module 103 is configured to perform fuzzy quantization processing on the error signal sample, the error signal trend parameter sample and the noise reduction signal sample respectively to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector; a third determining module 104 configured to determine a fuzzy control mapping relationship among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector; a fourth determining module 105 configured to determine the noise reduction parameters based on the error signal sample blur vector, the error signal trend parameter sample blur vector, the noise reduction signal sample blur vector, and the blur control mapping relationship.

In one embodiment, the fuzzy control mapping is used to eliminate the amplitude of the error signal samples while suppressing amplitude overshoot, ultimately bringing the amplitude to a state of minimum convergence.

Fig. 11 is a schematic structural diagram of a noise reduction parameter determination apparatus according to an embodiment of the present application. As shown in fig. 11, the second determining module 103 further includes: a first fuzzy partition unit 1031 configured to determine error signal fuzzy partition information and error signal trend parameter fuzzy partition information, respectively, with the error signal samples and the error signal trend parameter samples as observed quantities of fuzzy control; a first blur quantization unit 1032 configured to quantize the error signal sample blur into an error signal sample blur vector based on the error signal blur division information; a second fuzzy quantization unit 1033 configured to quantize the error signal trend parameter sample fuzzy into an error signal trend parameter sample fuzzy vector based on the error signal trend parameter fuzzy partition information; a second fuzzy partition unit 1034 configured to determine fuzzy partition information of the noise reduction signal by using the noise reduction signal sample as a control amount of fuzzy control; a third blur quantization unit 1035 configured to quantize the noise reduction signal sample blur into a noise reduction signal sample blur vector based on the noise reduction signal blur partition information.

In one embodiment, the first fuzzy partition unit 1031 is further configured to determine an error signal quantization level and an error signal domain, and an error signal trend parameter quantization level and an error signal trend parameter domain, respectively, with the error signal samples and the error signal trend parameter samples as observed quantities of fuzzy control; determining the membership degree of the error signal based on the quantization grade of the error signal and the discourse domain of the error signal, wherein the membership degree of the error signal is used for representing the degree of the accurate value of the discourse domain of the error signal which is subordinated to the quantization grade of the error signal; and determining the membership degree of the trend parameter of the error signal based on the quantization grade of the trend parameter of the error signal and the trend parameter domain of the error signal, wherein the membership degree of the trend parameter of the error signal is used for representing the degree of the accurate value of the trend parameter domain of the error signal which is subordinate to the quantization grade of the trend parameter of the error signal.

In one embodiment, the second fuzzy partition unit 1034 is further configured to determine the quantization level of the noise reduction signal and the discourse domain of the noise reduction signal by using the noise reduction signal samples as the control quantity of the fuzzy control; and determining the membership degree of the noise reduction signal based on the quantization level of the noise reduction signal and the domain of the noise reduction signal, wherein the membership degree of the noise reduction signal is used for representing the degree of the accurate value of the domain of the noise reduction signal which is subordinate to the quantization level of the noise reduction signal.

In one embodiment, the third determining module 104 is further configured to determine the quantization level of the noise reduction signal corresponding to the blur vector of the noise reduction signal sample to be positive if the quantization level of the error signal corresponding to the blur vector of the error signal sample is negative large or negative small and the quantization level of the error signal trend parameter corresponding to the blur vector of the error signal trend parameter sample is negative large or negative small; if the quantization level of the error signal corresponding to the fuzzy vector of the error signal sample is negative zero or positive zero, and the quantization level of the trend parameter of the error signal corresponding to the fuzzy vector of the trend parameter sample of the error signal is negative large or negative small, the quantization level of the noise reduction signal corresponding to the fuzzy vector of the noise reduction signal sample is positive small; and if the error signal quantization level corresponding to the error signal sample fuzzy vector is negative zero or positive zero, and the error signal trend parameter quantization level corresponding to the error signal trend parameter sample fuzzy vector is zero, the noise reduction signal quantization level corresponding to the noise reduction signal sample fuzzy vector is zero.

In one embodiment, the fourth determining module 105 is further configured to, for each fuzzy control mapping relationship in the plurality of fuzzy control mapping relationships, perform cartesian product operation on the error signal sample fuzzy vector and the noise reduction signal sample fuzzy vector which are mapped to each other corresponding to the fuzzy control mapping relationship to obtain a first intermediate value matrix corresponding to the fuzzy control mapping relationship, perform cartesian product operation on the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector which are mapped to each other corresponding to the fuzzy control mapping relationship to obtain a second intermediate value matrix corresponding to the fuzzy control mapping relationship, and perform minimum operation on elements in the first intermediate value matrix and the second intermediate value matrix to obtain a third intermediate value matrix corresponding to the fuzzy control mapping relationship; and carrying out maximum operation on elements of the third intermediate value matrix corresponding to the fuzzy control mapping relations respectively to obtain the noise reduction parameters.

The implementation process of the function and the effect of each module in the noise reduction parameter determination apparatus is specifically described in the implementation process of the corresponding step in the noise reduction parameter determination method, and is not described herein again.

Exemplary active noise reduction device

Fig. 12 is a schematic structural diagram of an active noise reduction device according to an embodiment of the present application. As shown in fig. 12, the active noise reduction device 200 includes: a fifth determining module 201, configured to determine a current error signal trend parameter corresponding to the current error signal based on the current error signal acquired by the error microphone, where the current error signal trend parameter is used to represent a subsequent operation trend of the current error signal; a fuzzy quantization module 202 configured to perform fuzzy quantization processing on the current error signal and the current error signal trend parameter to obtain a current error signal fuzzy vector and a current error signal trend parameter fuzzy vector; the fuzzy inference module 203 is configured to determine a current noise reduction signal fuzzy vector corresponding to the current noise reduction signal based on the current error signal fuzzy vector, the current error signal trend parameter fuzzy vector and the noise reduction parameter, where the noise reduction parameter is determined based on the noise reduction parameter determination method provided in any of the embodiments above; and the deblurring quantization module 204 is configured to perform deblurring quantization processing on the current noise reduction signal fuzzy vector to obtain a current noise reduction signal, so that the speaker plays the current noise reduction signal.

In one embodiment, the deblurring quantization module 204 is further configured to determine a current noise reduction signal universe to which the current noise reduction signal fuzzy vector is subordinate by searching for a maximum degree of membership; and determining the current noise reduction signal corresponding to the current noise reduction signal domain based on the corresponding relation between the noise reduction signal domain and the noise reduction signal.

The implementation process of the function and the effect of each module in the active noise reduction device is specifically described in the implementation process of the corresponding step in the active noise reduction method, and is not described herein again.

Exemplary electronic device

Fig. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 13, the electronic device 300 includes one or more processors 310 and memory 320.

The processor 310 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 300 to perform desired functions.

Memory 320 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium and executed by the processor 310 to implement the noise reduction parameter determination method or the active noise reduction method of the various embodiments of the present application described above and/or other desired functions.

In one example, the electronic device 300 may further include: an input device 330 and an output device 340, which are interconnected by a bus system and/or other form of connection mechanism (not shown).

Of course, for simplicity, only some of the components of the electronic device 300 relevant to the present application are shown in fig. 3, and components such as buses, input/output interfaces, and the like are omitted. In addition, electronic device 300 may include any other suitable components depending on the particular application.

Exemplary computer program product and computer-readable storage Medium

In addition to the above-described methods and apparatus, embodiments of the present application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps in the noise reduction parameter determination methods provided according to the various embodiments of the present application described in the above-mentioned "exemplary noise reduction parameter determination methods" section of this specification, or the steps in the active noise reduction methods provided according to the various embodiments of the present application described in the above-mentioned "exemplary active noise reduction methods" section.

The computer program product may write program code for carrying out operations for embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.

Furthermore, embodiments of the present application may also be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform the steps in the noise reduction parameter determination methods provided according to the various embodiments of the present application described in the "exemplary noise reduction parameter determination methods" section above or the steps in the active noise reduction methods provided according to the various embodiments of the present application described in the "exemplary active noise reduction methods" section above.

The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

It should be noted that the above listed embodiments are only specific examples of the present application, and obviously, the present application is not limited to the above embodiments, and many similar variations exist. All modifications which would occur to one skilled in the art and which are, therefore, directly derivable or suggested by the disclosure herein are to be included within the scope of the present application.

It should be understood that the terms first, second, etc. used in the embodiments of the present application are only used for clearly describing the technical solutions of the embodiments of the present application, and are not used to limit the protection scope of the present application.

The above description is only for the preferred embodiment of the present application, and is not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for determining noise reduction parameters, comprising:

acquiring a noise reduction signal sample output by a loudspeaker and an error signal sample acquired by an error microphone;

determining error signal trend parameter samples corresponding to the error signal samples based on the error signal samples, wherein the error signal trend parameter samples are used for representing subsequent running trends of the error signal samples;

respectively carrying out fuzzy quantization processing on the error signal sample, the error signal trend parameter sample and the noise reduction signal sample to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector;

determining a fuzzy control mapping relation among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector;

and determining noise reduction parameters based on the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector, the noise reduction signal sample fuzzy vector and the fuzzy control mapping relation.

2. The method according to claim 1, wherein the performing fuzzy quantization processing on the error signal samples, the error signal trend parameter samples and the noise reduction signal samples to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector respectively comprises:

respectively determining error signal fuzzy partition information and error signal trend parameter fuzzy partition information by taking the error signal sample and the error signal trend parameter sample as observed quantities of fuzzy control;

quantizing the error signal sample ambiguity into the error signal sample ambiguity vector based on the error signal ambiguity partition information;

based on the fuzzy dividing information of the error signal trend parameter, fuzzily quantizing the sample of the error signal trend parameter into a fuzzy vector of the sample of the error signal trend parameter;

determining fuzzy division information of the noise reduction signal by taking the noise reduction signal sample as a control quantity of fuzzy control;

and based on the fuzzy partition information of the noise reduction signal, carrying out fuzzy quantization on the noise reduction signal samples into fuzzy vectors of the noise reduction signal samples.

3. The noise reduction parameter determination method according to claim 2,

the method for determining the fuzzy division information of the error signals and the fuzzy division information of the trend parameters of the error signals respectively by taking the error signal samples and the trend parameter samples of the error signals as observed quantities of fuzzy control comprises the following steps:

respectively determining an error signal quantization grade and an error signal discourse domain, and an error signal trend parameter quantization grade and an error signal trend parameter discourse domain by taking the error signal sample and the error signal trend parameter sample as observed quantities of fuzzy control;

determining an error signal membership degree based on the error signal quantization grade and the error signal discourse domain, wherein the error signal membership degree is used for representing the degree of the accurate value of the error signal discourse domain subordinate to the error signal quantization grade;

determining a degree of membership of the trend parameter of the error signal based on the quantization level of the trend parameter of the error signal and the domain of the trend parameter of the error signal, wherein the degree of membership of the trend parameter of the error signal is used for representing the degree of membership of the accurate value of the domain of the trend parameter of the error signal to the quantization level of the trend parameter of the error signal;

and/or the presence of a gas in the gas,

the determining the fuzzy division information of the noise reduction signal by taking the noise reduction signal sample as the control quantity of fuzzy control comprises the following steps:

determining the quantization grade of the noise reduction signal and the discourse domain of the noise reduction signal by taking the noise reduction signal sample as the control quantity of fuzzy control;

and determining the membership degree of the noise reduction signal based on the quantization level of the noise reduction signal and the domain of the noise reduction signal, wherein the membership degree of the noise reduction signal is used for representing the degree of the accurate value of the domain of the noise reduction signal which is subordinate to the quantization level of the noise reduction signal.

4. The method according to any of claims 1 to 3, wherein the fuzzy control mapping is used to eliminate the amplitude of the error signal samples while suppressing the amplitude overshoot, and finally to keep the amplitude in a state of minimized convergence.

5. The method of claim 4, wherein the determining a fuzzy control mapping relationship between the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector, and the noise reduction signal sample fuzzy vector comprises:

if the quantization level of the error signal corresponding to the fuzzy vector of the error signal sample is negative large or negative small, and the quantization level of the trend parameter of the error signal corresponding to the fuzzy vector of the trend parameter sample of the error signal is negative large or negative small, the quantization level of the noise reduction signal corresponding to the fuzzy vector of the noise reduction signal sample is positive large;

if the quantization level of the error signal corresponding to the fuzzy vector of the error signal sample is negative zero or positive zero, and the quantization level of the trend parameter of the error signal corresponding to the fuzzy vector of the trend parameter sample of the error signal is negative large or negative small, the quantization level of the noise reduction signal corresponding to the fuzzy vector of the noise reduction signal sample is positive small;

and if the error signal quantization level corresponding to the error signal sample fuzzy vector is negative zero or positive zero, and the error signal trend parameter quantization level corresponding to the error signal trend parameter sample fuzzy vector is zero, the noise reduction signal quantization level corresponding to the noise reduction signal sample fuzzy vector is zero.

6. The method for determining noise reduction parameters according to any of claims 1 to 3, wherein said determining noise reduction parameters based on said error signal sample blur vector, said error signal trend parameter sample blur vector, said noise reduction signal sample blur vector and said blur control mapping relation comprises:

aiming at each fuzzy control mapping relation in a plurality of fuzzy control mapping relations, carrying out Cartesian product operation on error signal sample fuzzy vectors and noise reduction signal sample fuzzy vectors which are mapped with each other and correspond to the fuzzy control mapping relations to obtain a first intermediate value matrix corresponding to the fuzzy control mapping relations, carrying out Cartesian product operation on error signal trend parameter sample fuzzy vectors and noise reduction signal sample fuzzy vectors which are mapped with each other and correspond to the fuzzy control mapping relations to obtain a second intermediate value matrix corresponding to the fuzzy control mapping relations, and carrying out minimum operation on elements in the first intermediate value matrix and the second intermediate value matrix to obtain a third intermediate value matrix corresponding to the fuzzy control mapping relations;

and performing maximum operation on elements of a third intermediate value matrix corresponding to the fuzzy control mapping relations respectively to obtain the noise reduction parameters.

7. An active noise reduction method, comprising:

determining a current error signal trend parameter corresponding to a current error signal based on the current error signal acquired by an error microphone, wherein the current error signal trend parameter is used for representing a subsequent operation trend of the current error signal;

carrying out fuzzy quantization processing on the current error signal and the current error signal trend parameter to obtain a current error signal fuzzy vector and a current error signal trend parameter fuzzy vector;

determining a current noise reduction signal fuzzy vector corresponding to a current noise reduction signal based on the current error signal fuzzy vector, the current error signal trend parameter fuzzy vector and a noise reduction parameter, wherein the noise reduction parameter is determined based on the noise reduction parameter determination method according to any one of claims 1 to 6;

and carrying out deblurring quantization processing on the current noise reduction signal fuzzy vector to obtain the current noise reduction signal so that a loudspeaker can play the current noise reduction signal.

8. The active noise reduction method according to claim 7, wherein the deblurring and quantizing the current error signal blur vector to obtain the current noise reduction signal comprises:

determining a current noise reduction signal discourse domain to which the current noise reduction signal fuzzy vector is subjected by searching for the maximum membership degree;

and determining the current noise reduction signal corresponding to the domain of the current noise reduction signal based on the corresponding relation between the domain of the noise reduction signal and the noise reduction signal.

9. A noise reduction parameter determination apparatus, characterized by comprising:

the acquisition module is configured to acquire a noise reduction signal sample output by the loudspeaker and an error signal sample acquired by the error microphone;

a first determination module configured to determine an error signal trend parameter sample corresponding to the error signal sample based on the error signal sample, wherein the error signal trend parameter sample is used for characterizing a subsequent running trend of the error signal sample;

the second determining module is configured to perform fuzzy quantization processing on the error signal sample, the error signal trend parameter sample and the noise reduction signal sample respectively to obtain an error signal sample fuzzy vector, an error signal trend parameter sample fuzzy vector and a noise reduction signal sample fuzzy vector;

a third determining module configured to determine a fuzzy control mapping relationship among the error signal sample fuzzy vector, the error signal trend parameter sample fuzzy vector and the noise reduction signal sample fuzzy vector;

a fourth determining module configured to determine a noise reduction parameter based on the error signal sample blur vector, the error signal trend parameter sample blur vector, the noise reduction signal sample blur vector, and the blur control mapping relationship.

10. An active noise reduction device, comprising:

a fifth determining module, configured to determine a current error signal trend parameter corresponding to a current error signal based on the current error signal acquired by an error microphone, where the current error signal trend parameter is used to characterize a subsequent operation trend of the current error signal;

the fuzzy quantization module is configured to perform fuzzy quantization processing on the current error signal and the current error signal trend parameter to obtain a current error signal fuzzy vector and a current error signal trend parameter fuzzy vector;

a fuzzy inference module configured to determine a current noise reduction signal fuzzy vector corresponding to a current noise reduction signal based on the current error signal fuzzy vector, the current error signal trend parameter fuzzy vector and a noise reduction parameter, wherein the noise reduction parameter is determined based on the noise reduction parameter determination method according to any one of claims 1 to 6;

and the de-blurring quantization module is configured to perform de-blurring quantization processing on the current noise reduction signal fuzzy vector to obtain the current noise reduction signal, so that a loudspeaker plays the current noise reduction signal.