CN112260662A - Adaptive filtering method, computer equipment and device - Google Patents

Abstract

The application discloses a self-adaptive filtering method, which comprises the following steps: acquiring a current frame input signal, a current frame reference signal and a current frame filter coefficient; combining the current frame reference signal with the historical frame reference signal to obtain a current frame reference signal set; performing adaptive filtering on a current frame input signal based on a current frame reference signal and a current frame filter coefficient, and updating the current frame filter coefficient according to a preset division length based on a current frame reference signal set and the current frame input signal to obtain a filter coefficient of a subsequent frame; the current frame reference signal set and the historical frame reference signal set contain at least two parts with the same data, and the product of the number of interval frames between the historical frame and the current frame and the division length is equal to the product of the difference of the at least two parts and the current frame input signal length. By means of the method, the filter device and the filter method, the calculated amount in the filtering process can be prevented from being uneven, and the filtering effect is improved.

Description

A kind of adaptive filtering method, computer equipment and device

technical field

The present application relates to the field of filtering, and in particular, to an adaptive filtering method, computer equipment and apparatus.

Background technique

As we all know, audio communication has become a common method of information exchange in people's work and life. When people use a mobile phone to turn on a hands-free phone or a video conferencing terminal for a video conference, the sound played by the speaker will be picked up by the microphone again due to the external sound, which will cause echoes. The echo signal will seriously affect the communication quality and reduce the accuracy of speech recognition, so it is necessary to take effective measures to suppress the echo and improve the communication quality.

Usually, an adaptive filtering algorithm (echo canceling algorithm) is used to filter the collected signal to cancel the echo signal. During the long-term research and development process, the inventor of the present application found that the adaptive filtering algorithm, especially the algorithm for updating the filter coefficients, has the problem of uneven calculation amount, for example, the calculation amount increases sharply. When the load of the computing system is too large, it is easy to cause the failure of the adaptive filtering algorithm and even the collapse of the system.

SUMMARY OF THE INVENTION

The main technical problem to be solved by the present application is to provide an adaptive filtering method, computer equipment and device, which can avoid the uneven calculation amount in the filtering process and improve the filtering effect.

In order to solve the above-mentioned technical problems, a technical solution adopted in the present application is to provide an adaptive filtering method, including: obtaining the current frame input signal, the current frame reference signal and the current frame filter coefficient; The signal is combined to obtain the current frame reference signal set; the current frame input signal is adaptively filtered based on the current frame reference signal and the current frame filter coefficient, and the current frame filter coefficient is determined according to the preset based on the current frame reference signal set and the current frame input signal. The set division length is updated to obtain the filter coefficient of the subsequent frame; wherein, the current frame reference signal set and the historical frame reference signal set contain at least two parts with the same data, and the interval frame number between the historical frame and the current frame is the same as the division. The product of the lengths is equal to the product of the difference between the at least two parts and the length of the input signal of the current frame.

Wherein, combining the current frame reference signal and the historical frame reference signal to obtain the current frame reference signal set includes: combining the current frame reference signal and the historical frame reference signal according to the preset update signal block length to obtain a plurality of current frames The reference signal is divided into blocks; the current frame reference signal set includes a plurality of current frame reference signal blocks; wherein, the length of the update signal block is determined by the division length and the length of the current frame input signal.

Wherein, the length of the update signal block is not less than the sum of the division length and the length of the input signal of the current frame minus 1.

Among them, the length of the update signal block is equal to the power of 2.

The step of combining the current frame reference signal and the historical frame reference signal according to the preset update signal block length to obtain a plurality of current frame reference signal blocks includes: dividing the current frame filter coefficients according to the division lengths to obtain A plurality of current frame filter coefficient blocks, wherein the current frame filter coefficient blocks are in one-to-one correspondence with the current frame reference signal blocks; the current frame filter coefficient blocks and the corresponding current frame reference signal blocks are convolved to obtain a volume product signal; based on the difference between the convolution signal of the current frame and the input signal of the current frame, update the filter coefficient of the current frame to obtain the filter coefficient of the subsequent frame after the current frame.

Wherein, the reference signal sets of the current frame and the historical frame include at least two reference signal sub-blocks with the same data, which are the first reference signal sub-block and the second reference signal sub-block respectively; The product with the division length is equal to the product of the difference between the second reference signal block and the first reference signal block and the length of the input signal of the current frame.

Wherein, the division length is half the length of the input signal of each frame; or the division length is twice the length of the input signal of each frame.

The step of combining the current frame reference signal and the historical frame reference signal according to the preset update signal block length to obtain multiple current frame reference signal blocks includes: when the current frame input signal length is 160 and the division length is 80 , combine the current frame input reference signal and the historical frame reference signal with a length of 256 to obtain multiple current frame reference signal blocks; or when the current frame input signal length is 160 and the division length is 320, the current frame input reference signal The signal and the historical frame reference signal are combined with a length of 512 to obtain multiple current frame reference signal blocks.

In order to solve the above technical problem, another technical solution adopted in the present application is to provide a computer device including a processor for executing instructions to implement the above adaptive filtering method.

In order to solve the above technical problem, another technical solution adopted in the present application is to provide a device with a storage function, the device stores a program, and the above-mentioned adaptive filtering method can be implemented when the program is executed.

The beneficial effects of the present application are: different from the situation in the prior art, the present application provides an adaptive filtering method, by setting the division length in the update of the filter coefficient, and making the division length and the current frame input signal length satisfy constraints relation. This makes it possible to calculate all the input signals of the current frame in the updating process of each reference coefficient, thereby avoiding the retention of part of the signal data. Furthermore, the calculation amount of each filter coefficient update is equal, which avoids the problem of uneven calculation amount, and can effectively improve the update effect of the update algorithm.

Description of drawings

1 is a schematic flowchart of an adaptive filtering method according to an embodiment of the present application;

2 is a schematic flowchart of a filter coefficient update method according to an embodiment of the present application;

3 is a schematic flowchart of a method for determining a division length according to an embodiment of the present application;

4 is a schematic structural diagram of an adaptive filter according to an embodiment of the present application;

5 is a schematic structural diagram of a computer device according to an embodiment of the present application;

FIG. 6 is a schematic structural diagram of a device with a storage function according to an embodiment of the present application.

Detailed ways

In order to make the objectives, technical solutions and effects of the present application clearer and clearer, the present application will be further described in detail below with reference to the accompanying drawings and examples.

The embodiments of the present application can be used for echo cancellation processing of audio in an audio device, where the audio device may include, but is not limited to, a mobile communication device, a computer, a microphone, and the like. It should be understood that the application scenarios of the system and method of the present application are only some examples or embodiments of the present application. For those of ordinary skill in the art, without creative work, they can also use these drawings according to the drawings. Apply this application to other similar scenarios, such as adaptive filtering of video or adaptive filtering of other data.

Adaptive Filtering Algorithms Adaptive algorithms can be used to change filter coefficients or structures according to changes in the environment. Among them, the filter coefficients can be automatically and continuously adapted to the given signal to obtain the desired response. The updating method of the filter coefficients in the adaptive filtering method may include: Least Mean Square (LMS), Recursive Least Square (RLS), Normalized Least Square (Normalized Least Square) MeanSquares, NLMS) or Affine Projection Algorithm (Affine Projection Algorithm, APA). In the update process, the filter coefficient needs to be divided into a plurality of filter coefficient blocks before calculation.

For example, the length of the input signal of each frame may be 160, and the division length of the filter coefficients may be 128. Then in the update process, each time the block is calculated, the input signal has only 128 data for calculation, and 32 data points remain. When the number of stuck points reaches 128, two calculations will be performed at that moment, namely the calculation of the input signal of the current frame and the calculation of the stuck points. At this point, the amount of computation will be doubled. If the algorithm of the adaptive filtering method is relatively complex, when the budget increases sharply, the calculation of all data cannot be completed within the specified time, resulting in the failure of the algorithm and even the collapse of the system.

To solve the above problems, the present application discloses an adaptive filtering method. Referring to FIG. 1 , FIG. 1 is a schematic flowchart of an adaptive filtering method according to an embodiment of the present application. It should be noted that, if there is substantially the same result, the present embodiment is not limited to the sequence of the processes shown in FIG. 1 . As shown in Figure 1, the method includes:

Step S110: Obtain the current frame input signal, the current frame reference signal and the current frame filter coefficient.

In an embodiment, the current frame input signal and the current frame reference signal may refer to the input signal and the reference signal acquired in the current time period.

The input signal can be collected and transmitted through an audio capture device. Among them, the audio transmission is generally carried out in a frame of 10ms or 20ms. For example, when the sampling rate of the input audio signal is 16k, the discrete points of a frame of input signal data block are 160 points or 320 points, in this case, the length of each frame of input signal can be recorded as 160.

The reference signal may be obtained from the network side. The reference signal used for adaptive filtering in different scenarios will be different. For example, in the echo cancellation process, the reference signal may be the sound signal of the previous frame played by the speaker.

The filter coefficients in the adaptive filtering method are continuously updated during the filtering process. In one embodiment, the current frame filter coefficient used when processing the current frame input signal may be the filter coefficient updated according to the previous frame input signal and the previous frame reference signal. The filter coefficient can be a one-dimensional matrix containing N pieces of data.

Step S120: Combine the current frame reference signal and the historical frame reference signal to obtain the current frame reference signal set.

In one embodiment, the acquired current frame reference signal and the saved historical frame reference signal are combined to obtain the current frame reference signal set. The historical frame reference signal refers to a reference signal obtained at a moment before the current frame. The historical frame may refer to the previous moment of the current frame, or any moment before the current frame.

Each frame of reference signal has its corresponding reference signal set.

Step S130: Perform adaptive filtering on the current frame input signal based on the current frame filter coefficient and the current frame reference signal.

In one embodiment, the filtering process may be as follows: using the current frame filter coefficient and the current frame reference signal to perform convolution to obtain a filtered convolution signal. The filtered convolution signal is subtracted from the current frame input signal, that is, the filtered signal.

In one embodiment, the filtered signal can be sent to a speaker or a network terminal for output.

Step S140: Based on the current frame reference signal set and the current frame input signal, the filter coefficient of the current frame is updated according to the preset division length to obtain the filter coefficient of the subsequent frame after the current frame.

In one embodiment, the filter coefficient update method may include obtaining the division length; and updating the current frame filter coefficient based on the division length. The division length and the length of the input signal obtained by the adaptive filtering method of each frame satisfy the constraints. The constraint condition may refer to a relationship that needs to be satisfied between the division length and the length of the input signal of each frame. The division length can be preset or obtained from elsewhere. Obtaining filter coefficients of subsequent frames after the current frame may refer to obtaining filter coefficients at a moment after the current moment.

In one embodiment, the constraints may be: the current frame reference signal set and the historical frame reference signal set contain at least two parts with the same data, and the product of the number of interval frames and the division length between the historical frame and the current frame is equal to at least two parts. The product of the difference of the parts and the length of the input signal of the current frame. Specifically, the current frame reference signal set and the historical frame reference signal set at N moments before the current moment (eg, the previous moment, the previous two moments, etc.) include at least two parts with the same data. The product of the division length and N is equal to the product of the difference between these two parts and the length of the input signal of the current frame. The difference between the two parts refers to the difference between the numbers of the two parts in the corresponding reference signal set. The input signal length refers to the amount of data contained in each frame of the input signal.

In the adaptive filtering method disclosed in the present application, the division length in the updating of the filter coefficient is set, and the division length and the length of the input signal of the current frame satisfy the constraint relationship. This makes it possible to calculate all the input signals of the current frame in the updating process of each reference coefficient, thereby avoiding the retention of part of the signal data. Furthermore, the calculation amount of each filter coefficient update is equal, which avoids the problem of uneven calculation amount, and can effectively improve the update effect of the update algorithm.

The essence of the filter coefficient update process is to find a finite impulse response, so that the difference between the convolution signal obtained by convolving the response with the reference signal and the target echo signal is the smallest. That is, the echo signal is the smallest in the filtered output signal. Specifically, you can refer to formula (1):

为脉冲响应，即滤波系数。Where: d(i) is the input signal, x(i) is the reference signal,

is the impulse response, that is, the filter coefficient.

Referring to FIG. 2, FIG. 2 is a schematic flowchart of a method for updating a filter coefficient according to an embodiment of the present application. It should be noted that, if there is substantially the same result, the present embodiment is not limited to the sequence of the processes shown in FIG. 2 . As shown in Figure 2, the method includes:

Step S210: Obtain the division length.

In an embodiment, the division length P may be preset, or may be obtained by other modules or the network. The division length P may be a positive integer greater than zero. The division length P and the length L of the input signal obtained by the adaptive filtering method of each frame satisfy the constraints.

For more details of the method for determining the division length P, please refer to FIG. 3 and its related descriptions.

Step S220: Obtain the update signal block length.

Each time the filter coefficient is updated, the signal block length M required to be calculated is a power of 2. In an embodiment, the block length M of the update signal may refer to the number of data points in the data block of the reference signal for which the update calculation is performed, or may refer to the number of data points of the data block of the input signal for which the update calculation is performed.

In an embodiment, the update signal block length M may be determined based on the division length P and the input data length L of each frame. Specifically, the update signal block length M needs to satisfy the condition M≥P+L-1.

Step S230: Obtain multiple reference signal blocks.

In an embodiment, the current frame reference signal and the saved historical frame reference signal may be combined according to the update signal block length M. Thus, a plurality of reference signal blocks for filter coefficient update calculation can be obtained. Wherein, the amount of data in each reference signal block is equal to the length M of the update signal block.

The current frame reference signal set includes a plurality of current frame reference signal blocks.

Step S240: Obtain a plurality of filter coefficient blocks.

In an embodiment, the filter coefficients may be divided based on the division length P, thereby obtaining several filter coefficient blocks. See formula (2) for details:

为滤波系数分块，N为滤波系数的长度。in,

is the filter coefficient block, and N is the length of the filter coefficient.

The larger the data amount N of the filter coefficients, the better the filtering effect. However, in the filter coefficient update process, each update calculation needs to convolve the filter coefficient with the reference signal, and then compare it with the input signal, so the data amount of the filter coefficient in the update calculation will be limited. By setting the division length, the filter coefficient is divided into a plurality of filter coefficient blocks and then updated and calculated, so that the data amount of the filter coefficient is not limited above.

In one embodiment, the filter coefficient blocks and the reference signal blocks are in a one-to-one correspondence. That is, each filter coefficient block has its corresponding reference signal block.

Step S250: Perform Fourier transform on the filter coefficient blocks and the reference signal blocks.

In the method of updating the filter coefficients, the data can be converted from the time domain to the frequency domain first, and then the calculation is performed. The transformation of a signal from the time domain to the frequency domain can be accomplished by Fourier transform.

In one embodiment, the filter coefficient blocks are subjected to Fourier transform to obtain frequency domain filter coefficient blocks. Since the length of the signal block to be transformed is M, the Fourier transform matrix can be an M*M matrix. The data volume of the filter coefficient block will be less than the length M of the update signal block, so when performing Fourier transform, the method of supplementing 0 vectors can be used to make the data volume of the frequency domain phase filter coefficient block meet the update signal block length M. See formula (3) for details:

Among them, w _p is the frequency domain phase filter coefficient block, F represents the Fourier transform matrix of M*M, and 0 is 1-(MP) zero vectors.

In an embodiment, the reference signal block is subjected to Fourier transform, and the reference signal block vector is converted into a matrix, so that the convolution calculation can be performed with the filter coefficient block. See formula (4) for details:

X _p =diag{F[x[(n+1)L-pP-M+1],...,x[(n+1)L-pP]]} (4)

Wherein, X _P is the frequency domain filter coefficient block matrix; diag is the diagonal function; x[(n+1)L-pP] is the reference signal data numbered "(n+1)L-pP".

Step S260: Obtain a convolution signal.

其中I_L为L*L个单位向量。通过设置第一转换矩阵，可以将计算的数据中的部分数据去除，从而提高运算效率。In one embodiment, a signal transformation matrix may be constructed to remove redundant data or erroneous data in the calculation matrix, and only perform calculation on part of the data. Specifically, the first transformation matrix is

where _IL is L*L unit vectors. By setting the first transformation matrix, part of the data in the calculated data can be removed, thereby improving the operation efficiency.

In an embodiment, a convolution calculation may be performed on the effective filter coefficient blocks and the effective reference signal blocks to obtain a convolution signal. And, the obtained convoluted signal needs to be transformed from the frequency domain to the time domain. Specifically, the convolution signal matrix can refer to formula (5):

Step S270: Construct input signal blocks.

In one embodiment, an input signal block with M data may be constructed by adding a 0 vector. For details, see formula (6):

d=[0 d _n ] ^T (6)

Among them, d is the input signal block; d _n is the input signal of the nth frame, see formula (7) for details.

d _n =[d[nL+1],...,d[(n+1)L]] (7)

Step S280: Use the convolution signal and the input signal to block, and update the filter coefficient.

In one embodiment, the input signal block is subtracted from the convolution signal matrix to obtain the difference e, see formula (8):

e=d-y (8)

In one embodiment, the difference value may be subjected to Fourier transform, multiplied by the processed reference signal in blocks, and then added to the filter coefficient of the current frame to obtain the next filter coefficient. See formula (9) for details:

为下一滤波系数，

为当前帧滤波系数，F为傅里叶变换矩阵，

为X_p的共轭矩阵，G为约束变换矩阵，Δ为归一化系数。具体地，G和Δ参见公式(10)和(11)：in

is the next filter coefficient,

is the filter coefficient of the current frame, F is the Fourier transform matrix,

is the conjugate matrix of X _p , G is the constraint transformation matrix, and Δ is the normalization coefficient. Specifically, G and Δ refer to equations (10) and (11):

Δ=1/(βΔ+(1-β)|X _p | ² ) (11)

为第二转换矩阵。第二转换矩阵可以用于去掉计算矩阵中的冗余数据或错误数据，仅对部分数据进行计算。Among them, β is a smoothing parameter, which can be taken as 0.92.

is the second transformation matrix. The second transformation matrix can be used to remove redundant data or erroneous data in the calculation matrix, and only perform calculation on part of the data.

It can be seen from the above update method of the filter coefficient that the division length of the filter coefficient will directly affect the calculation amount of the update process. Referring to FIG. 3 , FIG. 3 is a schematic flowchart of a method for determining a division length according to an embodiment of the present application. It should be noted that, if there is substantially the same result, the present embodiment is not limited to the sequence of the processes shown in FIG. 3 . As shown in Figure 3, the method includes:

Step S310: Determine the constraint relationship.

In order for the convolution operation to be effective, the condition M≥P+L-1 needs to be satisfied. Meanwhile, in order to improve the operation efficiency, the data (eg, X _p ) calculated when the reference signal of the current frame is updated can be reused in the subsequent update of the reference coefficient, so that the amount of operation can be reduced.

In an embodiment, the reference signal sets of the current frame and the historical frame include at least two reference signal blocks with the same data, which are the first reference signal block and the second reference signal block, respectively. That is, the aforementioned reference signal set of the current frame and the reference signal set of the historical frame containing at least two parts with the same data may refer to at least two reference signal blocks with the same data. The reference signal data used in the first reference signal sub-block and the second reference signal sub-block needs to satisfy the conditions as in the formula (12) before they can be reused in the filter coefficient update of multiple frames.

Wherein, n1 is the number of frames by which the filter coefficient of the historical frame differs from the filter coefficient of the current frame.

Formula (12) can be understood as the p1th reference signal when the filter coefficient of the current frame (ie, the n+ _n1th frame) is updated and the p2th reference signal when the filter coefficient before the _n1th frame (ie, the nth frame) is updated. The item reference signal blocks are the same.

Meanwhile, p1 and p2 also need to satisfy p ₁ >p ₂ .

By simplifying the formula (12), the constraint relationship between the division length and the input signal length can be obtained. Constraints are: the product of the number of interval frames between the historical frame and the current frame and the division length is equal to the product of the difference between the first reference signal block and the second reference signal block and the length of the input signal of the current frame. See equation (13):

n ₁ L=(p ₁ -p ₂ )P (13)

Step S320: Determine the degree of reuse of reference signal blocks.

In one embodiment, the degree of reuse may be determined in advance according to the actual application scenario or the configuration of the computing device. Specifically, the number of frames between the historical frame and the current frame with the largest number of reference signal blocks that can be reused (for example, n1), and the number of reference signal blocks that can be reused (for example, the value of p1-p2) can be determined. ).

Step S330: Determine the division length and update the signal block length.

In one embodiment, the previously determined reusability of the reference signal is substituted into the constraint relationship, and the division length of the filter coefficient can be obtained based on the length of the input signal. In addition, the value range of M can be determined according to the condition M≥P+L-1. And since M needs to be a power of 2, the value of M can be a power of 2 that is closest to P+L-1.

For example, when setting p ₁ -p ₂ =2, n ₁ =1, that is, the reference signal block used for updating the filter coefficients of the current frame is different from the reference signal block used for updating the filter coefficients of the previous historical frame only by two items . The length L of the input signal of each frame may be 160, and the above value is substituted into the constraint relationship to obtain L=2P, that is, the division length P may be 80. The value of M needs to satisfy 3P+1=241<M, and M can be 256. To sum up, when the current frame input signal length is 160 and the division length is 80, the current frame input reference signal and the historical frame reference signal are combined with a length of 256 to obtain multiple current frame reference signal blocks. In this scheme, a single update of the filter coefficients can substitute all the input signals of the current frame, thereby effectively solving the problem of data retention.

For another example, when setting p ₁ -p ₂ =1, n ₁ =2, there is only one reference signal block used for updating the filter coefficients of the current frame and the reference signal block used for updating the filter coefficients of the second frame before the current frame Are not the same. The length L of the input signal of each frame can be 160, and the above value is substituted into the constraint relationship to obtain 2L=P, that is, P can be 320. The value of M needs to satisfy 3L-1=479<M, and M can be 512. To sum up, when the length of the current frame input signal is 160 and the division length is 320, the current frame input reference signal and the historical frame reference signal are combined with a length of 512 to obtain multiple current frame reference signal blocks. In this scheme, a single update of the filter coefficients can substitute all the input signals of the current frame, thereby effectively solving the problem of data retention. Also, the computational complexity of the entire adaptive filter can be reduced.

Please refer to the drawings. FIG. 4 is a schematic structural diagram of an adaptive filter according to an embodiment of the present application. In this embodiment, the adaptive filter 400 includes an acquiring module 410 , a combining module 420 , a filtering module 430 and an updating module 440 .

In one embodiment, the obtaining module 410 is configured to obtain the current frame input signal, the current frame reference signal and the current frame filter coefficient.

In one embodiment, the combining module 420 is configured to combine the current frame reference signal and the historical frame reference signal to obtain the current frame reference signal set. The combining module 420 may also be configured to combine the current frame reference signal and the historical frame reference signal to obtain the current frame reference signal set, which includes: combining the current frame reference signal and the historical frame reference signal according to the preset update signal block length, to obtain a set of current frame reference signals. A plurality of current frame reference signal blocks are obtained; the current frame reference signal set includes a plurality of current frame reference signal blocks; wherein, the length of the update signal block is determined by the division length and the length of the current frame input signal. The length of the update signal block is not less than the sum of the division length and the length of the input signal of each frame minus 1. The update signal block length is equal to a power of 2.

In one embodiment, the filtering module 430 is configured to perform adaptive filtering on the current frame input signal based on the current frame filter coefficient and the current frame reference signal.

In one embodiment, the update module 440 is configured to perform adaptive filtering on the current frame input signal based on the current frame reference signal and the current frame filter coefficient, and the current frame filter coefficient is preset based on the current frame reference signal set and the current frame input signal. The division length is updated to obtain the filter coefficient of the subsequent frame; wherein, the current frame reference signal set and the historical frame reference signal set contain at least two parts with the same data, the interval frame number between the historical frame and the current frame and the division length The product of is equal to the product of the difference between at least two parts and the length of the input signal of the current frame.

The update module 440 can also be used to divide the filter coefficients of the current frame according to the division length to obtain a plurality of filter coefficient blocks of the current frame, wherein the filter coefficient blocks of the current frame correspond to the reference signal blocks of the current frame one-to-one; The filter coefficient blocks and the corresponding current frame reference signal blocks are convolved to obtain a convolution signal; based on the difference between the current frame convolution signal and the current frame input signal, the current frame filter coefficient is updated. Wherein, the reference signal sets of the current frame and the historical frame include at least two reference signal sub-blocks with the same data, which are the first reference signal sub-block and the second reference signal sub-block respectively; The product with the division length is equal to the product of the difference between the second reference signal block and the first reference signal block and the length of the input signal of the current frame. Wherein, the division length is half the length of the input signal of each frame; or the division length is twice the length of the input signal of each frame. The step of combining the current frame reference signal and the historical frame reference signal according to the preset update signal block length to obtain multiple current frame reference signal blocks includes: when the current frame input signal length is 160 and the division length is 80 , combine the current frame input reference signal and the historical frame reference signal with a length of 256 to obtain multiple current frame reference signal blocks; or when the current frame input signal length is 160 and the division length is 320, the current frame input reference signal The signal and the historical frame reference signal are combined with a length of 512 to obtain multiple current frame reference signal blocks.

Please refer to FIG. 5 , which is a schematic structural diagram of a computer device according to an embodiment of the present application. In this embodiment, computer device 500 includes processor 510 .

The processor 510 may also be referred to as a CPU (Central Processing Unit, central processing unit). The processor 510 may be an integrated circuit chip with signal processing capability. Processor 510 may also be a general purpose processor, digital signal processor (DSP), application specific integrated circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components . A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

Computer device 500 may further include a memory (not shown) for storing instructions and data required by processor 510 to operate.

The processor 510 is configured to execute the instructions to implement the method provided by any of the above-mentioned adaptive filtering method embodiments and any non-conflicting combination.

Please refer to FIG. 6 , which is a schematic structural diagram of a device with a storage function according to an embodiment of the present application. The device 600 with the storage function of the embodiment of the present application stores an instruction, and when the instruction is executed, implements the method provided by any embodiment of the adaptive filtering method of the present application and any non-conflicting combination. Wherein, the instruction can be stored in the above-mentioned device with a storage function in the form of a program file in the form of a software product, so that a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) executes this instruction. All or part of the steps of the method of each embodiment are claimed. The aforementioned device 600 with a storage function includes: a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, etc. The medium of program code, or terminal equipment such as computers, servers, mobile phones, and tablets.

In the several embodiments provided in this application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, for example, multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The above description is only an embodiment of the present application, and is not intended to limit the scope of the patent of the present application. Any equivalent structure or equivalent process transformation made by using the contents of the description and drawings of the present application, or directly or indirectly applied to other related technologies Fields are similarly included within the scope of patent protection of this application.

Claims (10)

1. An adaptive filtering method, comprising:

acquiring a current frame input signal, a current frame reference signal and a current frame filter coefficient;

combining the current frame reference signal with a historical frame reference signal to obtain a current frame reference signal set;

performing adaptive filtering on the current frame input signal based on the current frame reference signal and the current frame filter coefficient, and updating the current frame filter coefficient according to a preset division length based on the current frame reference signal set and the current frame input signal to obtain a filter coefficient of a subsequent frame;

the current frame reference signal set and the historical frame reference signal set comprise at least two parts with the same data, and the product of the number of interval frames between the historical frame and the current frame and the division length is equal to the product of the difference of the at least two parts and the length of the current frame input signal.

2. The adaptive filtering method according to claim 1, wherein said combining the current frame reference signal with a historical frame reference signal to obtain a current frame reference signal set comprises:

combining the current frame reference signal and the historical frame reference signal according to a preset updating signal block length to obtain a plurality of current frame reference signal blocks; the current frame reference signal set comprises a plurality of the current frame reference signal blocks;

wherein the update signal block length is determined by the division length and the current frame input signal length.

3. The adaptive filtering method according to claim 2, wherein the update signal block length is not less than a value obtained by subtracting 1 from a sum of the division length and the current frame input signal length.

4. The adaptive filtering method according to claim 3, wherein the update signal block length is equal to a power of 2.

5. The adaptive filtering method according to claim 2, wherein the step of combining the current frame reference signal with the historical frame reference signal according to a preset update signal block length to obtain a plurality of current frame reference signal blocks is followed by:

dividing the current frame filter coefficient according to the division length to obtain a plurality of current frame filter coefficient blocks, wherein the current frame filter coefficient blocks correspond to the current frame reference signal blocks one to one;

convolving the current frame filter coefficient block and the corresponding current frame reference signal block to obtain a convolution signal;

and updating the current frame filter coefficient based on the difference between the current frame convolution signal and the current frame input signal to obtain the filter coefficient of a subsequent frame after the current frame.

6. The adaptive filtering method according to claim 5,

the reference signal sets of the current frame and the historical frame comprise at least two reference signal blocks with the same data, namely a first reference signal block and a second reference signal block;

the product of the number of interval frames between the historical frame and the current frame and the division length is equal to the product of the difference value of the second reference signal block and the first reference signal block and the input signal length of the current frame.

7. The adaptive filtering method according to claim 1,

the division length is one half of the length of each frame of input signals; or

The division length is twice the length of the input signal of each frame.

8. The adaptive filtering method according to claim 2, wherein the step of combining the current frame reference signal and the historical frame reference signal according to a preset update signal block length to obtain a plurality of current frame reference signal blocks comprises:

when the length of the current frame input signal is 160 and the division length is 80, combining the current frame input reference signal and the historical frame reference signal by 256 lengths to obtain a plurality of current frame reference signal blocks; or

When the current frame input signal length is 160 and the division length is 320, combining the current frame input reference signal and the historical frame reference signal with 512 lengths to obtain a plurality of current frame reference signal blocks.

9. A computer device comprising a processor for executing instructions to implement the adaptive filtering method of any one of claims 1-8.

10. An apparatus having a memory function, wherein the apparatus stores a program, which when executed is capable of implementing the adaptive filtering method according to any one of claims 1 to 8.

Images