CN113655436B - Method and device for optimizing broadband wave beam formation through channel calibration particle swarm - Google Patents
Method and device for optimizing broadband wave beam formation through channel calibration particle swarm Download PDFInfo
- Publication number
- CN113655436B CN113655436B CN202110839718.XA CN202110839718A CN113655436B CN 113655436 B CN113655436 B CN 113655436B CN 202110839718 A CN202110839718 A CN 202110839718A CN 113655436 B CN113655436 B CN 113655436B
- Authority
- CN
- China
- Prior art keywords
- frequency
- calibration
- compensation
- weight coefficient
- fir filter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000002245 particle Substances 0.000 title claims abstract description 96
- 238000000034 method Methods 0.000 title claims abstract description 77
- 230000015572 biosynthetic process Effects 0.000 title description 3
- 238000005457 optimization Methods 0.000 claims abstract description 59
- 239000011159 matrix material Substances 0.000 claims abstract description 49
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 42
- 230000001629 suppression Effects 0.000 claims abstract description 30
- 230000003044 adaptive effect Effects 0.000 claims abstract description 21
- 238000001914 filtration Methods 0.000 claims abstract description 12
- 238000001228 spectrum Methods 0.000 claims abstract description 11
- 108091006146 Channels Proteins 0.000 claims description 59
- 239000013598 vector Substances 0.000 claims description 52
- 238000005070 sampling Methods 0.000 claims description 17
- 230000003595 spectral effect Effects 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 6
- 230000003321 amplification Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 4
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 5
- 230000015556 catabolic process Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000006731 degradation reaction Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 230000005764 inhibitory process Effects 0.000 description 2
- 230000005236 sound signal Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000005311 autocorrelation function Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000005316 response function Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
- G01S3/80—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using ultrasonic, sonic or infrasonic waves
- G01S3/8003—Diversity systems specially adapted for direction finding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Remote Sensing (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Radar, Positioning & Navigation (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The invention discloses a method and a device for forming an optimized broadband beam with a band-pass calibration particle swarm. The method comprises the following steps: calculating compensation coefficients of the adaptive calibration compensation filters corresponding to the microphone array element channels; determining a frequency band range of interest to a user and main lobes and side lobes of an array beam on the frequency band range; calculating an input spectrum power autocorrelation matrix on a frequency band range of interest of a user, and constructing a problem of solving the weight coefficient of the optimal FIR filter under a linear constraint minimum variance criterion; converting the problem into an unconstrained optimization problem, and solving the problem by using a particle swarm optimization algorithm to obtain a weight coefficient of an optimal FIR filter; summing the digital signals subjected to filtering compensation by using the weight coefficient of the optimal FIR filter to output time domain signals subjected to beam forming; if the incoming wave direction or side lobe constraint, interference and noise suppression requirements of the target sound source pointed by the wave beam are changed, the weight coefficient of the FIR filter is recalculated.
Description
Technical Field
The invention relates to a wideband beam forming method for optimizing a band-pass band calibration particle swarm, and also relates to a corresponding wideband beam forming device, belonging to the technical field of acoustics.
Background
With the development of array signal processing technology, a beam forming (beam forming) method is widely applied to the fields of radar, communication, intelligent voice and the like. Array speech enhancement applications require the formation of a relatively uniform wideband beam over the audio frequency range, increasing array directivity, effectively enhancing the target speech signal while suppressing interference and various random noise.
The broadband beam forming method can be divided into a fixed beam forming method and an adaptive beam forming method, and the adaptive beam forming method fully utilizes the statistical characteristics of the array received signals, but in practical application, particularly when the number of array elements is large, the statistical characteristics are difficult to acquire in real time, the algorithm operation amount is too large, and the application of the adaptive beam forming method is limited. The fixed beam forming method does not need to update operation in real time, and the beam is designed on a wide frequency band according to the array characteristics and the performance index requirements, so that the practical value is higher.
The prior art mainly adopts a time domain broadband fixed beam former, and forms expected broadband beams meeting constraint conditions by optimizing the weight coefficient of each channel filter formed by the beams. The beam former generally adopts a time domain broadband beam forming method with side lobe control, converts the problem of solving the beam forming weight coefficient into a convex optimization problem, sets main lobe broadband constraint and side lobe suppression constraint on a broadband frequency band, searches the optimal solution of the convex optimization problem, and can obtain array broadband beams with good space directivity and interference noise suppression. However, this method has the following disadvantages: firstly, the types of interference and noise in the actual acoustic environment are complex, and the influence of factors such as scattered noise, strong directional interference sources and the like exists, so that the array processing performance is reduced; secondly, the synthesized beam is not accurate enough; thirdly, the robustness of the method is poor, and the mismatch of amplitude or phase among array elements can cause serious degradation of the array beam performance, thereby seriously affecting the practical degree.
Disclosure of Invention
The primary technical problem to be solved by the invention is to provide a wideband beam forming method optimized by using a band-pass calibration particle swarm.
Another technical problem to be solved by the present invention is to provide a wideband beam forming device with a band pass calibration particle swarm optimization.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
according to a first aspect of an embodiment of the present invention, there is provided a method for optimizing broadband beamforming with a band pass calibration particle swarm, comprising the steps of:
step S1, performing self-adaptive calibration in a calibration stage, and calculating compensation coefficients of self-adaptive calibration compensation filters corresponding to each microphone array element channel;
step S2, determining a frequency band range of interest to a user and main lobes and side lobes of an array beam on the frequency band range;
s3, calculating an input spectrum power autocorrelation matrix on a frequency band range of interest of a user, and constructing a problem of solving the weight coefficient of the optimal FIR filter under a linear constraint minimum variance criterion;
s4, converting the problem into an unconstrained optimization problem, and solving the problem by using a particle swarm optimization algorithm to obtain a weight coefficient of an optimal FIR filter;
s5, summing the digital signals subjected to filtering compensation by the adaptive calibration compensation filter by adopting the weight coefficient of the optimal FIR filter, and outputting a time domain signal subjected to beam forming; and if the incoming wave direction of the target sound source pointed by the wave beam or the side lobe constraint, interference and noise suppression requirements are not changed, repeatedly executing the step S5, otherwise, returning to the step S2, and recalculating the weight coefficient of the FIR filter.
Preferably, the step S2 comprises the following sub-steps:
s11, acquiring a played broadband calibration signal by adopting a microphone array;
and S12, taking the first microphone array element channel as a reference channel, calculating error signal target values of self-adaptive calibration between the rest microphone array element channels and the reference channel, and calculating compensation coefficients of the self-adaptive calibration compensation filters corresponding to the microphone array element channels by adopting a proportional normalization least mean square algorithm.
Preferably, in step S12, the error signal target value of the adaptive calibration between each remaining microphone element channel and the reference channel is expressed as:
|e n (m)| 2 ＝|x 0 (m)-s n (m)g n (m)| 2
in the above, e n (m) represents the digital signal x output by the adaptive calibration compensation filter corresponding to the nth microphone element channel n (m) and reference target signal x 0 (m) errors between; broadband calibration signal s collected by nth microphone array element n (m)＝[s n (m),...,s n (m-M+1)] T M represents the sampling point of the first microphone array element channel, and M represents the sampling point delayQuantity of time g n (m) represents the compensation coefficient corresponding to the current sampling point.
Preferably, when the proportional normalized least mean square algorithm calculates the compensation coefficient of the adaptive calibration compensation filter corresponding to each microphone array element channel, an iteration formula is adopted as follows:
in the above, g n (m+1) represents the compensation coefficient corresponding to the next sampling point, g n (m) represents a compensation coefficient corresponding to the current sampling point, eta represents a corrected step constant, delta is a smaller integer, and the stability reduction caused by the overlarge step constant eta due to the overlarge inner product of an input vector is prevented; t represents a transpose; g (m+1) =diag { β 1 (m+1),β 2 (m+1),...,β M (m+1) } is a step control matrix, diag { } represents a diagonal matrix, and β (m+1) represents an element in the step control matrix.
Wherein preferably, the element beta (m+1) in the step control matrix is obtained according to the following formula;
compensation coefficient feedback value ζ l (m+1)＝max{ρmax{υ,|g 1 (m)|,...,|g M (m)|},|g l (m) | } prevents the compensation coefficient from being too small to cause iteration failure, v is a correction value for preventing the compensation coefficient from being zero, and ρ is a compensation coefficient feedback scale factor.
Preferably, the step S2 comprises the following sub-steps:
s21, determining a frequency band range of interest of a user, dividing the frequency band range into K+1 frequency points, selecting R+1 reference frequency points from the frequency points, and calculating frequency values of related frequency points;
step S22, determining the reference beam width corresponding to the frequency reference value of each reference frequency point, and obtaining the beam main lobe width corresponding to the frequency value of all the frequency points in the frequency band range of interest to the user by carrying out interpolation operation on the reference beam width;
and S23, determining side lobe areas and main lobe beams corresponding to the frequency values of all frequency points in the frequency band range of interest of the user according to the width of the main lobe of the beam, and applying constraint to fine division of the side lobe areas to obtain side lobe constraint direction values on all frequency points.
Preferably, the step S3 comprises the following substeps:
step S31, dividing N again in the frequency band of interest to the user f +1 frequency points, and calculating the frequency value of each frequency point;
step S32, respectively calculating a target sound source, uncorrelated background noise, scattered noise components and a spectral power autocorrelation matrix of directional interference;
and step S33, summing up the target sound source, the uncorrelated background noise, the scattered noise component and the spectral power autocorrelation matrix of the directional interference, performing regularization treatment to obtain an input spectral power autocorrelation matrix, and constructing a problem of solving the weight coefficient of the optimal FIR filter under the linear constraint minimum variance criterion.
Preferably, in step S33, the following constraint condition is satisfied, and the problem of solving the weight coefficient w of the optimal FIR filter under the linear constraint minimum variance criterion is constructed:
constraint conditions
In the above, V s ＝[v s (0),v s (1),...,v s (K)]A matrix of K +1 frequency bin target signal steering vectors representing divisions in a frequency band range of interest to the user,represents a K+1-dimensional distortion-free constraint vector, f k Representing the frequency value corresponding to each frequency point in K+1 frequency points, f s Representation ofDigital audio sampling frequency; />Represents a matrix formed by the E side lobe constraint guide vectors of K+1 frequency points, the expression euclidean norms, epsilon sl And epsilon n Parameters of the side lobe suppression degree and the white noise gain amplification are shown respectively.
Preferably, the step S4 comprises the following substeps:
step S41, converting the constructed problem of the weight coefficient of the optimal FIR filter meeting the constraint condition into an unconstrained optimization problem:
，
wherein Ω (w) represents an objective function of the optimization problem, w represents a weight coefficient of the FIR filter, R x Representing the input spectral power autocorrelation matrix, N sl Representing the number of side lobe constraint, and lambda represents a positive weight coefficient;
step S42, setting related parameters of a particle swarm optimization algorithm, initializing each population particle in a solution space, and setting a position vector and a speed vector of each population particle to be respectivelyAnd->Np is the population particle number, Q is the particle sequence number, and t is the iteration number of the particle swarm optimization algorithm;
step S43, orderSubstituting the position vectors of the particles of each population into the unconstrained optimization problem to calculate an objective function +.>And searching and finding the self history of population particlesOptimal solution vectorAnd historical optimal solution vectors for all population particlest' represents the iteration number of the particle swarm optimization algorithm;
step S44, when the iteration number T of the particle swarm optimization algorithm reaches the preset maximum iteration number T max Terminating the iteration of the particle swarm optimization algorithm to obtain the weight coefficient of the optimal FIR filterAnd executing step S5; otherwise, updating the speed vector and the position vector of the particle swarm, and returning to the step S43 to continue iteration.
Preferably, in step S44, when the iteration number T of the particle swarm optimization algorithm does not reach the preset maximum iteration number T max When the method is used, the optimal solution vector of the population particles is calculated according to the iterationAnd->Updating the velocity vector and the position vector of the particle swarm:
in the above, the inertia factor omega and the learning factorSelection of [0,1 ] according to the particular application]Number of intervals.
According to a second aspect of an embodiment of the present invention, there is provided a wideband beamforming apparatus with band pass calibration particle swarm optimization, comprising a processor and a memory, the processor reading a computer program or instructions in the memory for performing the following operations:
step S1, performing self-adaptive calibration in a calibration stage, and calculating compensation coefficients of self-adaptive calibration compensation filters corresponding to each microphone array element channel;
step S2, determining a frequency band range of interest to a user and main lobes and side lobes of an array beam on the frequency band range;
s3, calculating an input spectrum power autocorrelation matrix on a frequency band range of interest of a user, and constructing a problem of solving the weight coefficient of the optimal FIR filter under a linear constraint minimum variance criterion;
s4, converting the problem into an unconstrained optimization problem, and solving the problem by using a particle swarm optimization algorithm to obtain a weight coefficient of an optimal FIR filter;
s5, summing the digital signals subjected to filtering compensation by the adaptive calibration compensation filter by adopting the weight coefficient of the optimal FIR filter, and outputting a time domain signal subjected to beam forming; and if the incoming wave direction of the target sound source pointed by the wave beam or the side lobe constraint, interference and noise suppression requirements are not changed, repeatedly executing the step S5, otherwise, returning to the step S2, and recalculating the weight coefficient of the FIR filter.
Compared with the prior art, the method and the device for forming the optimized broadband beam of the band-pass calibration particle swarm provided by the invention have the following characteristics:
(1) The beam design with the beam control and the complex interference noise suppression is considered, the restraint is implemented on the strong directional interference, the scattered noise and the white noise gain in a targeted manner, and the strong directional interference can be obviously suppressed and the output signal to noise ratio can be improved under the condition that the requirements of the main lobe width and the side lobe level are met.
(2) The particle swarm optimization method is adopted for solving the optimization problem, the optimal solution of the weight coefficient of the FIR filter can be rapidly and accurately obtained, the expected wave beam is accurately synthesized, and the wave beam directivity and the interference noise suppression capability are effectively improved.
(3) In the calibration stage, the compensation coefficient of the calibration compensation filter is obtained by utilizing the self-adaptive filtering architecture and the proportional normalization least mean square algorithm, the array beam performance degradation caused by poor consistency among microphone array element channels is eliminated, the effectiveness and the robustness of the technical method are obviously improved, and the method has more practical value.
Drawings
Fig. 1 is a flowchart of optimizing broadband beam forming by using a band pass calibration particle swarm according to an embodiment of the present invention;
fig. 2 is a schematic block diagram of a wideband beamforming method for optimizing a band pass calibration particle swarm according to an embodiment of the present invention;
fig. 3 is a schematic block diagram of adaptive calibration compensation in a wideband beamforming method with band pass calibration particle swarm optimization provided in an embodiment of the present invention;
fig. 4 is a schematic diagram comparing beam patterns of the wideband beamforming method optimized by the band pass calibration particle swarm and other beamforming methods according to the embodiment of the present invention;
FIG. 5 is a schematic diagram comparing a beam pattern of a band pass calibration particle swarm optimized wideband beam forming method and a band pass calibration beam forming method according to an embodiment of the present invention;
fig. 6a to fig. 6c are schematic diagrams of comparison between a band pass calibration particle swarm optimized wideband beamforming method and an output signal waveform and a spectrogram of an existing beamforming method according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a wideband beamforming device with band pass calibration particle swarm optimization according to an embodiment of the present invention.
Detailed Description
The technical contents of the present invention will be described in further detail with reference to the accompanying drawings and specific examples.
In order to obtain a multi-channel filtering weight coefficient for beam forming and realize an optimal expected waveform meeting constraint conditions, the space directivity, the specific interference noise suppression capability and the robustness of the beam are obviously improved, and as shown in fig. 1, the embodiment of the invention provides a wideband beam forming method for optimizing a band-pass calibration particle swarm, which comprises the following steps:
step S1, performing self-adaptive calibration in a calibration stage, and calculating compensation coefficients of self-adaptive calibration compensation filters corresponding to each microphone array element channel.
The method comprises the following substeps:
and S11, acquiring a played broadband calibration signal by adopting a microphone array.
In the embodiment of the invention, a linear microphone array consisting of N omnidirectional microphone array elements is adopted to collect and play broadband calibration signals. The linear microphone array is distributed on a z coordinate axis of a rectangular coordinate system, each microphone array element is used as a microphone array element channel, the microphone array element channel is represented by a serial number N, wherein n=0, 1d represents the spacing of the microphone elements.
In the calibration phase, the played broadband calibration signal can be a voice or broadband noise signal, and is used as a calibration input sound source, the incident direction of the input sound source is at a vertical incidence angle of 90 degrees with the linear microphone array, and according to a far-field acoustic model, the sound signal can be considered to be transmitted to each microphone array element without delay.
And S12, taking the first microphone array element channel as a reference channel, calculating the self-adaptive calibration error signal target value between each remaining microphone array element channel and the reference channel, and calculating the compensation coefficient of the self-adaptive calibration compensation filter corresponding to each microphone array element channel by adopting a proportional normalization least mean square algorithm.
As shown in fig. 2 and 3, a wideband calibration signal s is acquired for each microphone array element n (m) after cascade connection of adaptive calibration compensation filters, the signals are fed into an FIR filter, and the input signal of the FIR filter is x n (m). Selecting the first microphone array element channel 0 as a reference channel, and outputting a digital signal x by the reference channel 0 (m) as a reference target signal, the impulse response function of the microphone element channel 0 is g 0 (M) =δ (M-M), M represents the sampling point of the first microphone element channel 0, M represents the number of sampling point delays, δ is a smaller integer, equal to the order of the adaptive calibration compensation filter, preventing the input vectorToo small an inner product causes a too large step factor eta to cause a decrease in stability.
The error signal target value of the adaptive calibration between each microphone array element channel and the reference channel is expressed as follows:
|e n (m)| 2 ＝|x 0 (m)-s n (m)g n (m)| 2 (1)
in the above, e n (m) represents the digital signal x output by the adaptive calibration compensation filter corresponding to the nth microphone element channel n (m) and reference target signal x 0 Error between (m), i.e. e n (m)＝x 0 (m)-x n (m). Broadband calibration signal s collected by nth microphone array element n (m)＝[s n (m),...,s n (m-M+1)] T . Calculating |e by adopting a proportional normalization least mean square algorithm n (m)| 2 G corresponding to the smallest hour n As compensation coefficients for the adaptive calibration compensation filter corresponding to the nth microphone element channel.
When calculating the compensation coefficient of the self-adaptive calibration compensation filter corresponding to each microphone array element channel by the proportional normalization least mean square algorithm, the adopted iterative formula is as follows:
in the above, g n (m+1) represents the compensation coefficient corresponding to the next sampling point, g n (m) represents the compensation coefficient corresponding to the current sampling point, eta represents the corrected step constant, e n (m) represents the digital signal x output by the adaptive calibration compensation filter corresponding to the nth microphone element channel n (m) and reference target signal x 0 The error between (m), delta is a smaller integer, preventing the stability from being reduced due to the fact that the step constant eta is too large because the inner product of the input vector is too small; T representing a transpose; g (m+1) =diag { β 1 (m+1),β 2 (m+1),...,β M (m+1) } is the step control matrix, diag { } represents the diagonal matrix, and β (m+1) represents the elements in the step control matrixThe element is calculated according to the following formula.
Compensation coefficient feedback value ζ l (m+1)＝max{ρmax{υ,|g 1 (m)|,...,|g M (m)|},|g l And (M) is smaller to prevent the compensation coefficient from being too small to cause iteration failure, v is a correction value for preventing the compensation coefficient from being zero, ρ is a compensation coefficient feedback scale factor, and ρ is generally equal to or more than 1/M and equal to or less than 5/M. The convergence rate of the proportional normalized least mean square algorithm is high, when |e n (m)| 2 < epsilon meets the iterative convergence condition, and after convergence, a fixed calibration filter coefficient g is obtained n N=0, 1,..n-1, compensates for errors due to mismatch of the transfer functions between channels. Where ε represents the mean square error threshold.
And S2, determining a frequency band range of interest to the user, and main lobes and side lobes of the array beam on the frequency band range.
The main lobe of the array beam in the frequency band of interest to the user comprises a beam main lobe width and a beam main lobe direction corresponding to the frequency values of all the frequency points, and a side lobe area and a side lobe constraint direction.
The method comprises the following substeps:
step S21, determining the frequency band range of interest of the user, dividing the frequency band range into K+1 frequency points, selecting R+1 reference frequency points from the frequency points, and calculating the frequency value of the related frequency points.
According to the requirements in practical application, determining the lower frequency limit f L And an upper frequency limit f H Constitutes the frequency band range of interest to the user [ f L ,f H ]. Wherein, satisfy 0 < f L ＜f H ≤f s /2，f s Digital audio sampling frequency.
In the frequency band range of interest to the user [ f L ,f H ]Among K+1 frequency points of the internal division, the frequency value f corresponding to each frequency point k The method comprises the following steps:for example, in the frequency band range of interest to the user [ f L ,f H ]120 frequency points are divided internally.
Uniformly selecting a frequency lower limit f from the divided K+1 frequency points L And an upper frequency limit f H R+1 reference frequency points of (2), K is usually required to be divided by R, frequency reference value f r The method comprises the following steps:for example, 5 frequency points are selected from the 120 frequency points divided as reference frequency points, respectively [ B (f) 0 )，B(f 1 )，B(f 2 )，B(f 3 )，B(f 4 )]＝[100°,80°,65°，60°,55°]。
Step S22, determining the frequency reference value f of each reference frequency point r Corresponding reference beam width B (f r ) And performing interpolation operation by referring to the beam width to obtain the beam main lobe width corresponding to the frequency values of all the frequency points in the frequency band range of interest of the user.
The beam width of the conventional method is taken as a basis, the main lobe beam is widened and narrowed according to the practical application requirement, and different frequency reference values f are determined r Corresponding reference beam width B (f r ). According to different frequency reference values f r Corresponding reference beam width B (f r ) Calculating the frequency values f of all frequency points in the frequency band range of interest by adopting an interpolation algorithm k Corresponding beam main lobe width B (f k ). The frequency value f of all the frequency points is obtained by a piecewise linear interpolation method k Corresponding beam main lobe width B (f k ). The calculation formula under the piecewise linear interpolation method is as follows:
step S23, according to the beam main lobe width corresponding to the frequency values of all the frequency points, a side lobe area and a main lobe beam corresponding to the frequency values of all the frequency points in the frequency band range interested by the user are determined, and the side lobe area is finely divided and restrained, so that side lobe constraint direction values on all the frequency points are obtained.
Knowing the frequency values f of all the frequency points in the frequency band of interest to the user k Corresponding beam main lobe width B (f k ) Then, the part outside the width of the main lobe of the beam is regarded as a side lobe area, and in order to obtain better side lobe suppression, the fine division of the side lobe area is restricted, so that N sl For the number of side lobe constraints, θ s For the target sound source direction angle, the sidelobe constraint direction values on all the frequency points are as follows:
in the above formula, v=0,.. sl -1，
In addition, since the beam main lobe direction follows the target sound source angle, the frequency value f according to all frequency points k The corresponding beam main lobe width and target sound source angle can also determine the frequency value f of all frequency points k The corresponding main lobe beam, specifically denoted as θ s +B(f k )/2[θ s -B(f k )/2，θ s +B(f k )/2]. For example, the frequency value f of a certain frequency point k The corresponding beam main lobe width is 20 degrees, the target sound source angle is 70-90 degrees, and then the frequency value f of the frequency point can be obtained k The corresponding main lobe direction of the beam is 80-100 degrees.
And S3, calculating an input spectrum power autocorrelation matrix on a frequency band range of interest of a user according to the expected beam design requirement, interference suppression and noise reduction requirement, and solving the problem of the weight coefficient of the optimal FIR filter under the linear constraint minimum variance criterion.
The method comprises the following substeps:
step S31, dividing N again in the frequency band of interest to the user f +1 frequency points, and frequency values of the respective frequency points are calculated.
In the frequency band range of interest to the user [ f L ,f H ]N of internal division f Of +1 frequency points, the frequency value f corresponding to each frequency point q The method comprises the following steps:
wherein N is f The larger the value is, the finer the frequency division is, the higher the calculation accuracy of the broadband input spectrum power autocorrelation matrix is, and the more favorable is for accurately obtaining the weight coefficient of the FIR filter meeting the requirement.
Step S32, calculating the spectral power autocorrelation matrixes of the target sound source, the uncorrelated background noise, the scattered noise component and the directional interference respectively.
If the target sound source direction angle theta is known s Its spectral power autocorrelation matrix R s Obtained according to the following formula.
In the above, sigma s The weighting coefficient representing the target sound source signal is in the range of 0,1]Re represents the real part, H represents the conjugate transpose, v s (q) is the frequency value f corresponding to the frequency point q Can be expressed as:
in the above formula, I represents an N x N-dimensional identity matrix,representing the kronecker product of the two,representing an lx1 dimensional vector, L representing the FIR filter order,represents an N x 1-dimensional vector, where j represents the imaginary part, d' = [ z ] 0 ,z 1 ,...,z N-1 ] T A coordinate vector representing the position of the microphone array, c representing the propagation velocity of sound in air.
Assuming that the background noise on each microphone array channel is uncorrelated and all satisfy a gaussian distribution, let the uncorrelated noise variance beThe spectral power autocorrelation matrix of the uncorrelated background noise is obtained according to the following formula.
In the above, I NL Represents a nl×nl dimensional identity matrix.
Assuming that the actual environment is a diffuse noise field,is the variance of the scattered noise, and the spectral power autocorrelation matrix R of the scattered noise component df Obtained according to the following formula.
In the above-mentioned method, the step of,is a toeplitz matrix, e q For the first column element of the matrix,/o>For the first row of elements of the matrix, sinc (·) is a sine function, D is an N-dimensional matrix representing the differential information of the positions of the microphone elements, and the elements { D }, in the U-th row and Y-th column of the matrix UY ＝|z U -z Y |，U,Y＝0,1,...,N-1。
Setting interference zero limit for the condition of strong directional interference signal, the spectrum power of directional interference is selfCorrelation matrix R B Obtained according to the following formula.
In the above, v F (q) is a guiding vector of interference directions, the set interference zero limit number is J, and the interference zero limit direction is theta F F=1, 2,..j, the corresponding interference variance parameter isd′＝[z 0 ,z 1 ,...,z N-1 ] T A coordinate vector representing the position of the microphone array.
And step S33, summing up the target sound source, the uncorrelated background noise, the scattered noise component and the spectral power autocorrelation matrix of the directional interference, performing regularization treatment to obtain an input spectral power autocorrelation matrix, and constructing a problem of solving the weight coefficient of the optimal FIR filter under the linear constraint minimum variance criterion.
The input spectral power autocorrelation matrix is the input spectral power autocorrelation matrix corresponding to the digital signal calibrated by the adaptive calibration compensation filter. The input spectrum power autocorrelation matrix is obtained by the following regularization processing formula.
R x ＝R s +R n +R df +R B +μI NL (12)
In the above equation, μ represents a regularization coefficient.
On the premise of guaranteeing the improvement of operation robustness, the following constraint conditions are met, and the problem of solving the weight coefficient w of the optimal FIR filter under the linear constraint minimum variance criterion is constructed:
constraint conditions
In the above, V s ＝[v s (0),v s (1),...,v s (K)]Representing the frequency band range of interest to the user [ f L ,f H ]The matrix formed by the guide vectors of the K+1 frequency point target signals divided in the middle,represents a K+1-dimensional distortion-free constraint vector, < >>Representing the frequency band range of interest to the user [ f L ,f H ]The E side lobe constraint guide vector of the K+1 frequency points divided in the middle, the expression euclidean norms, epsilon sl And epsilon n Parameters of the side lobe suppression degree and the white noise gain amplification are shown respectively.
And S4, converting the constructed problem of the weight coefficient of the optimal FIR filter meeting the constraint condition into an unconstrained optimization problem, and solving the problem by using a particle swarm optimization algorithm to obtain the weight coefficient of the optimal FIR filter.
The method comprises the following substeps:
step S41, converting the constructed problem of the weight coefficient of the optimal FIR filter meeting the constraint condition into an unconstrained optimization problem:
wherein Ω (w) is an objective function of the optimization problem, a weight coefficient w of the FIR filter that minimizes Ω (w) is calculated, λ being a positive weight coefficient.
Step S42, setting related parameters of a particle swarm optimization algorithm, initializing each population particle in a solution space, and setting a position vector and a speed vector of each population particle to be respectivelyAnd->Np is the population particle number, Q is the particle sequence number, and t is the iteration number of the particle swarm optimization algorithm.
In the embodiment of the invention, the particle swarm optimization algorithm is an optimization algorithm for searching an optimal solution through cooperation and information sharing among individuals in a swarm.
Setting the related parameters of the particle swarm optimization algorithm to comprise the iteration times T and the maximum iteration times T max Population particle count N p Inertia factor omega and learning factor omega are respectively
Let iteration number t=0, initialize N in solution space p And setting the position vector and the speed vector of NL dimension of each population particle as follows:and->Any one of the set position vectors is expressed as:
in the above, I K+1 Is a k+1-dimensional identity matrix, and α is a regularization parameter. The position vectors and velocity vectors for other populations of particles may be randomly generated in the solution space.
Step S43, orderSubstituting the position vectors of the particles of each population into the unconstrained optimization problem to calculate an objective function +.>Searching and finding historical optimal solution vector of population particleAnd the historical optimal solution vector of all population particles +.>t' represents the number of iterations of the particle swarm optimization algorithm.
Step S44, when the iteration number T of the particle swarm optimization algorithm reaches the preset maximum iteration number T max Terminating the iteration of the particle swarm optimization algorithm to obtain the weight coefficient of the optimal FIR filterAnd executing step S5; otherwise, updating the speed vector and the position vector of the particle swarm, and returning to the step S43 to continue iteration.
As shown in fig. 2, the weight coefficient w of the optimal FIR filter obtained by the particle swarm optimization algorithm is used as the weight coefficient of each FIR filter, and can be expressed as h n (m)，n＝0,1,...,N-1。
When the iteration times T of the particle swarm optimization algorithm do not reach the preset maximum iteration times T max When the method is used, the optimal solution vector of the population particles is calculated according to the iterationAnd->Updating the velocity vector and the position vector of the particle swarm:
in the above, the inertia factor omega and the learning factorSelected according to the specific applicationSelect [0,1]Number of intervals.
S5, adopting a weight coefficient of an optimal FIR filter to carry out summation on the digital signals subjected to filtering compensation by the self-adaptive calibration compensation filters corresponding to the microphone array element channels and outputting time domain signals subjected to beam forming; and if the incoming wave direction of the target sound source pointed by the wave beam or the side lobe constraint, interference and noise suppression requirements are not changed, repeatedly executing the step S5, otherwise, returning to the step S2, and recalculating the weight coefficient of the FIR filter.
As shown in fig. 2, the ambient sound signal s collected in real time by each microphone array element channel n (m) the digital signal x after being filtered and compensated by the calibration and compensation filter n (m) outputting the time domain signal y (m) after summation and beam forming to an FIR filter with the optimal weight coefficient.
Fig. 4 is a comparison of the beam patterns of the present invention with other beam forming methods. It can be seen that the beam sidelobes formed by the conventional frequency domain delay and sum beam forming method are higher, and the directivity and the robustness in the low frequency part are poor. The linear constraint minimum variance beam forming method adopts an FIR filtering addition structure, and the time domain weighting filtering is used for directly meeting the optimal criterion and constraint conditions in a wide frequency band range, so that the sidelobe suppression degree is improved by about 1-2 dB compared with the delay addition method. The improved linear constraint minimum variance beam forming method can flexibly control a sidelobe suppression degree by increasing sidelobe suppression constraint and constructing an input spectrum power autocorrelation function with directional interference and noise suppression on the basis of the original method, the sidelobe average suppression degree in the example is 20dB, strong suppression nulls can be formed in a specific direction, and meanwhile the problem of noise amplification is avoided. The method of the invention searches the optimal weight coefficient of the FIR filter by adopting the particle swarm optimization algorithm on the basis of the improved linear constraint minimum variance method, can more accurately approach the global optimal solution of the optimization problem while maintaining the advantages, has a side lobe average suppression degree of about 26dB and has stronger noise reduction capability.
Fig. 5 is a comparison of the beam patterns of the present invention with the non-channel calibration beam forming. Because of the random deviation of amplitude or phase among microphone array element channels, the sidelobe suppression degree of the channel-free calibration beam forming method is obviously reduced compared with the ideal case of consistent channels, and the sidelobe curve of the channel-free calibration beam forming method is very close to the sidelobe curve under each ideal case, and the sidelobe suppression degree can be improved by about 10 dB. The method can solve the problems of performance degradation and robustness caused by channel consistency by simply determining the compensation coefficient through the self-adaptive calibration process in the calibration stage.
Fig. 6 a-6 c are graphs comparing waveforms and spectrograms of signals output by the method of the present invention with those of the prior art beamforming method. FIG. 6a is a diagram showing an input signal including a target sound source signal, white noise, and two sets of directional interference signals, one set being a 115-degree directional speech signal and one set being a 40-degree directional mono signal; fig. 6b is an output of the prior art method, and it can be seen that, due to higher side lobe of the beam, the suppression of the beam on two sets of directional interference signals is relatively poor, and the residue of the output waveform interference signal is relatively high; fig. 6c shows that the method of the invention can reduce side lobes, and can form strong zero limit inhibition to the single-tone interference signal in a specific direction, so that the 115-degree voice signal in the strong zero limit inhibition direction can be almost completely inhibited, and the single-tone interference signal in the 40-degree side lobe direction has only weak residue, and in addition, the method of the invention has good noise reduction capability.
In addition, as shown in fig. 7, the embodiment of the present invention further provides a wideband beam forming device with a band pass calibration particle swarm optimization, which includes a processor 32 and a memory 31, and may further include a communication component, a sensor component, a power component, a multimedia component and an input/output interface according to actual needs. Wherein the memory, communication components, sensor components, power components, multimedia components, and input/output interfaces are all coupled to the processor 32. As mentioned above, the memory 31 may be a Static Random Access Memory (SRAM), an Electrically Erasable Programmable Read Only Memory (EEPROM), an Erasable Programmable Read Only Memory (EPROM), a Programmable Read Only Memory (PROM), a Read Only Memory (ROM), a magnetic memory, a flash memory, or the like; processor 32 may be a Central Processing Unit (CPU), a Graphics Processor (GPU), a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), a Digital Signal Processing (DSP) chip, or the like. Other communication components, sensor components, power components, multimedia components, etc. may be implemented using common components found in existing smartphones and are not specifically described herein.
In addition, the wideband beamforming device for band pass calibration particle swarm optimization provided by the embodiment of the invention comprises a processor 32 and a memory 31, wherein the processor 32 reads a computer program or instructions in the memory 31 and is used for executing the following operations:
step S1, performing self-adaptive calibration in a calibration stage, and calculating compensation coefficients of self-adaptive calibration compensation filters corresponding to each microphone array element channel.
And S2, determining a frequency band range of interest to the user, and main lobes and side lobes of the array beam on the frequency band range.
And S3, calculating an input spectrum power autocorrelation matrix on a frequency band range of interest of a user according to the expected beam design requirement, interference suppression and noise reduction requirement, and solving the problem of the weight coefficient of the optimal FIR filter under the linear constraint minimum variance criterion.
And S4, converting the constructed problem of the weight coefficient of the optimal FIR filter meeting the constraint condition into an unconstrained optimization problem, and solving the problem by using a particle swarm optimization algorithm to obtain the weight coefficient of the optimal FIR filter.
S5, adopting a weight coefficient of an optimal FIR filter to carry out summation on the digital signals subjected to filtering compensation by the self-adaptive calibration compensation filters corresponding to the microphone array element channels and outputting time domain signals subjected to beam forming; and if the incoming wave direction of the target sound source pointed by the wave beam or the side lobe constraint, interference and noise suppression requirements are not changed, repeatedly executing the step S5, otherwise, returning to the step S2, and recalculating the weight coefficient of the FIR filter.
In addition, the embodiment of the present invention further provides a computer readable storage medium, where instructions are stored on the readable storage medium, when the computer readable storage medium runs on a computer, to enable the computer to execute the method for optimizing broadband beamforming with a channel calibration particle swarm as described in fig. 1, and specific implementation manners thereof are not repeated herein.
In addition, the embodiment of the present invention further provides a computer program product containing instructions, which when executed on a computer, cause the computer to perform the method for optimizing wideband beamforming with channel calibration particles as described in fig. 1, and detailed implementation manner thereof will not be repeated here.
Compared with the prior art, the method and the device for forming the optimized broadband beam of the band-pass calibration particle swarm provided by the invention have the following characteristics:
(1) The beam design with the beam control and the complex interference noise suppression is considered, the restraint is implemented on the strong directional interference, the scattered noise and the white noise gain in a targeted manner, and the strong directional interference can be obviously suppressed and the output signal to noise ratio can be improved under the condition that the requirements of the main lobe width and the side lobe level are met.
(2) The particle swarm optimization method is adopted for solving the optimization problem, the optimal solution of the weight coefficient of the FIR filter can be rapidly and accurately obtained, the expected wave beam is accurately synthesized, and the wave beam directivity and the interference noise suppression capability are effectively improved.
(3) In the calibration stage, the compensation coefficient of the calibration compensation filter is obtained by utilizing the self-adaptive filtering architecture and the proportional normalization least mean square algorithm, the array beam performance degradation caused by poor consistency among microphone array element channels is eliminated, the effectiveness and the robustness of the technical method are obviously improved, and the method has more practical value.
The method and the device for optimizing the broadband beam forming through the band pass calibration particle swarm provided by the invention are described in detail. Any obvious modifications to the present invention, without departing from the spirit of the present invention, would be apparent to those skilled in the art from the scope of the present patent claims.
Claims (11)
1. The wideband beam forming method for optimizing the band-pass band calibration particle swarm is characterized by comprising the following steps of:
step S1, performing self-adaptive calibration in a calibration stage, and calculating compensation coefficients of self-adaptive calibration compensation filters corresponding to each microphone array element channel;
step S2, determining a frequency band range of interest to a user and main lobes and side lobes of an array beam on the frequency band range;
s3, calculating an input spectrum power autocorrelation matrix on a frequency band range of interest of a user, and constructing a problem of solving the weight coefficient of the optimal FIR filter under a linear constraint minimum variance criterion;
s4, converting the problem into an unconstrained optimization problem, and solving the problem by using a particle swarm optimization algorithm to obtain a weight coefficient of an optimal FIR filter;
s5, summing the digital signals subjected to filtering compensation by the adaptive calibration compensation filter by adopting the weight coefficient of the optimal FIR filter, and outputting a time domain signal subjected to beam forming; and if the incoming wave direction of the target sound source pointed by the wave beam or the side lobe constraint, interference and noise suppression requirements are not changed, repeatedly executing the step S5, otherwise, returning to the step S2, and recalculating the weight coefficient of the FIR filter.
2. The method for optimized wideband beamforming with band pass calibration particles as claimed in claim 1, wherein step S2 comprises the sub-steps of:
s11, acquiring a played broadband calibration signal by adopting a microphone array;
and S12, taking the first microphone array element channel as a reference channel, calculating error signal target values of self-adaptive calibration between the rest microphone array element channels and the reference channel, and calculating compensation coefficients of the self-adaptive calibration compensation filters corresponding to the microphone array element channels by adopting a proportional normalization least mean square algorithm.
3. The method for optimized wideband beamforming with band pass calibration particles as claimed in claim 2, wherein in step S12,
the error signal target value of the adaptive calibration between each microphone array element channel and the reference channel is expressed as:
|e n (m)| 2 ＝|x 0 (m)-s n (m)g n (m)| 2
in the above, e n (m) represents the digital signal x output by the adaptive calibration compensation filter corresponding to the nth microphone element channel n (m) and reference target signal x 0 (m) errors between; broadband calibration signal s collected by nth microphone array element n (m)＝[s n (m),...,s n (m-M+1)] T M represents the sampling point of the first microphone array element channel, M represents the delay quantity of the sampling point, g n (m) represents the compensation coefficient corresponding to the current sampling point.
4. A method of optimizing broadband beamforming with band pass calibration particles as claimed in claim 3, wherein:
when the proportional normalization least mean square algorithm calculates the compensation coefficient of the self-adaptive calibration compensation filter corresponding to each microphone array element channel, an iteration formula is adopted as follows:
in the above, g n (m+1) represents the compensation coefficient corresponding to the next sampling point, g n (m) represents a compensation coefficient corresponding to the current sampling point, eta represents a step constant, delta is a smaller integer, and the stability reduction caused by the overlarge step constant eta due to the overlarge inner product of an input vector is prevented; t represents a transpose; g (m+1) =diag { β 1 (m+1),β 2 (m+1),...,β M (m+1) } is a step control matrix, diag { } represents a diagonal matrix, and β (m+1) represents an element in the step control matrix.
5. The method for optimized wideband beamforming with band pass calibration particles of claim 4, wherein:
the element beta (m+1) in the step control matrix is obtained according to the following formula;
compensation coefficient feedback value ζ l (m+1)＝max{ρmax{υ,|g 1 (m)|,...,|g M (m)|},|g l And (m) | } is used for preventing the compensation coefficient from being too small to cause iteration failure, v is a correction value for preventing the compensation coefficient from being zero, and ρ is a compensation coefficient feedback scale factor.
6. The method for optimized wideband beamforming with band pass calibration particles as claimed in claim 1, wherein step S2 comprises the sub-steps of:
s21, determining a frequency band range of interest of a user, dividing the frequency band range into K+1 frequency points, selecting R+1 reference frequency points from the frequency points, and calculating frequency values of related frequency points;
step S22, determining the reference beam width corresponding to the frequency reference value of each reference frequency point, and obtaining the beam main lobe width corresponding to the frequency value of all the frequency points in the frequency band range of interest to the user by carrying out interpolation operation on the reference beam width;
and S23, determining side lobe areas and main lobe beams corresponding to the frequency values of all frequency points in the frequency band range of interest of the user according to the width of the main lobe of the beam, and applying constraint to fine division of the side lobe areas to obtain side lobe constraint direction values on all frequency points.
7. The method for optimized wideband beamforming with band pass calibration particles as recited in claim 1, wherein step S3 comprises the sub-steps of:
step S31, dividing N again in the frequency band of interest to the user f +1 frequency points, and calculating the frequency value of each frequency point;
step S32, respectively calculating a target sound source, uncorrelated background noise, scattered noise components and a spectral power autocorrelation matrix of directional interference;
and step S33, summing up the target sound source, the uncorrelated background noise, the scattered noise component and the spectral power autocorrelation matrix of the directional interference, performing regularization treatment to obtain an input spectral power autocorrelation matrix, and constructing a problem of solving the weight coefficient of the optimal FIR filter under the linear constraint minimum variance criterion.
8. The method for optimized wideband beamforming with band pass calibration particles as claimed in claim 7, wherein in step S33, the following constraint condition is satisfied, and the problem of solving the weight coefficient w of the optimal FIR filter under the linear constraint minimum variance criterion is constructed:
in the above, V s ＝[v s (0),v s (1),...,v s (K)]A matrix of K +1 frequency bin target signal steering vectors representing divisions in a frequency band range of interest to the user,represents a K+1-dimensional distortion-free constraint vector, f k Representing the frequency value corresponding to each frequency point in K+1 frequency points, f s Representing a digital audio sampling frequency; v (V) θsl,E ＝[v sl,E (0),v sl,E (1),...,v sl,E (K)]Represents a matrix formed by E side lobe constraint guide vectors of K+1 frequency points, and I I.I. represents Euclidean norms and epsilon sl And epsilon n Parameters respectively representing sidelobe suppression degree and white noise gain amplification, R x Representing the input spectral power autocorrelation matrix.
9. The method for optimized wideband beamforming with band pass calibration particles as recited in claim 8, wherein step S4 comprises the sub-steps of:
step S41, converting the constructed problem of the weight coefficient of the optimal FIR filter meeting the constraint condition into an unconstrained optimization problem:
，
wherein Ω (w) represents an objective function of the optimization problem, w represents a weight coefficient of the FIR filter, R x Representing the input spectral power autocorrelation matrix, N sl Representing the number of side lobe constraint, and lambda represents a positive weight coefficient;
step S42, setting related parameters of a particle swarm optimization algorithm, initializing each population particle in a solution space, and setting a position vector and a speed vector of each population particle to be respectivelyAnd->Np is the population particle number, Q is the particle sequence number, and t is the iteration number of the particle swarm optimization algorithm;
step S43, orderSubstituting the position vectors of the particles of each population into the unconstrained optimization problem to calculate an objective function +.>Searching and finding historical optimal solution vector of population particleAnd historical optimal solution vectors for all population particlest' represents the iteration number of the particle swarm optimization algorithm;
step S44, when the iteration number T of the particle swarm optimization algorithm reaches the preset maximum iteration number T max The particle swarm optimization algorithm is terminatedIterating to obtain the weight coefficient of the optimal FIR filterAnd executing step S5; otherwise, updating the speed vector and the position vector of the particle swarm, and returning to the step S43 to continue iteration.
10. The method of claim 9, wherein in step S44, when the iteration number T of the particle swarm optimization algorithm does not reach the preset maximum iteration number T max When the method is used, the optimal solution vector of the population particles is calculated according to the iterationAnd->Updating the velocity vector and the position vector of the particle swarm:
in the above, the inertia factor omega and the learning factorSelection of [0,1 ] according to the particular application]Number of intervals.
11. A band pass calibration particle swarm optimized wideband beam forming apparatus comprising a processor and a memory, said processor reading a computer program or instructions in said memory for performing the operations of:
step S1, performing self-adaptive calibration in a calibration stage, and calculating compensation coefficients of self-adaptive calibration compensation filters corresponding to each microphone array element channel;
step S2, determining a frequency band range of interest to a user and main lobes and side lobes of an array beam on the frequency band range;
s3, calculating an input spectrum power autocorrelation matrix on a frequency band range of interest of a user, and constructing a problem of solving the weight coefficient of the optimal FIR filter under a linear constraint minimum variance criterion;
s4, converting the problem into an unconstrained optimization problem, and solving the problem by using a particle swarm optimization algorithm to obtain a weight coefficient of an optimal FIR filter;
s5, summing the digital signals subjected to filtering compensation by the adaptive calibration compensation filter by adopting the weight coefficient of the optimal FIR filter, and outputting a time domain signal subjected to beam forming; and if the incoming wave direction of the target sound source pointed by the wave beam or the side lobe constraint, interference and noise suppression requirements are not changed, repeatedly executing the step S5, otherwise, returning to the step S2, and recalculating the weight coefficient of the FIR filter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110839718.XA CN113655436B (en) | 2021-07-23 | 2021-07-23 | Method and device for optimizing broadband wave beam formation through channel calibration particle swarm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110839718.XA CN113655436B (en) | 2021-07-23 | 2021-07-23 | Method and device for optimizing broadband wave beam formation through channel calibration particle swarm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113655436A CN113655436A (en) | 2021-11-16 |
| CN113655436B true CN113655436B (en) | 2024-04-16 |
Family
ID=78478115
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110839718.XA Active CN113655436B (en) | 2021-07-23 | 2021-07-23 | Method and device for optimizing broadband wave beam formation through channel calibration particle swarm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113655436B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114280544B (en) * | 2021-12-02 | 2023-06-27 | 电子科技大学 | Minimum transition band width direction diagram shaping method based on relaxation optimization |
| CN116611223B (en) * | 2023-05-05 | 2023-12-19 | 中国科学院声学研究所 | Accurate array response control method and device combined with white noise gain constraint |
| CN116683965B (en) * | 2023-07-20 | 2023-10-20 | 之江实验室 | Digital beam forming device, method and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108872870A (en) * | 2018-06-21 | 2018-11-23 | 浙江工业大学 | A kind of lithium battery SOC estimation method based on particle group optimizing expanded Kalman filtration algorithm |
| CN109959917A (en) * | 2019-03-08 | 2019-07-02 | 南京航空航天大学 | A kind of non-frequency in broadband becomes the array Sparse methods of multi-beam imaging sonar |
| CN111487594A (en) * | 2020-04-23 | 2020-08-04 | 中国民航大学 | Circular array beam forming method based on particle swarm optimization |
| CN112202483A (en) * | 2020-10-12 | 2021-01-08 | 西安工程大学 | Beam forming method and device, electronic equipment and storage medium |
-
2021
- 2021-07-23 CN CN202110839718.XA patent/CN113655436B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108872870A (en) * | 2018-06-21 | 2018-11-23 | 浙江工业大学 | A kind of lithium battery SOC estimation method based on particle group optimizing expanded Kalman filtration algorithm |
| CN109959917A (en) * | 2019-03-08 | 2019-07-02 | 南京航空航天大学 | A kind of non-frequency in broadband becomes the array Sparse methods of multi-beam imaging sonar |
| CN111487594A (en) * | 2020-04-23 | 2020-08-04 | 中国民航大学 | Circular array beam forming method based on particle swarm optimization |
| CN112202483A (en) * | 2020-10-12 | 2021-01-08 | 西安工程大学 | Beam forming method and device, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 粒子群优化算法在FIR数字滤波器设计中的应用;周飞红;刘辉;廖子贞;;计算机工程与应用(第33期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113655436A (en) | 2021-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113655436B (en) | Method and device for optimizing broadband wave beam formation through channel calibration particle swarm | |
| US8050717B2 (en) | Signal processing system and method for calibrating channel signals supplied from an array of sensors having different operating characteristics | |
| Benesty et al. | Conventional beamforming techniques | |
| US7626889B2 (en) | Sensor array post-filter for tracking spatial distributions of signals and noise | |
| CN107223345B (en) | FIR filter coefficient calculation for beamforming filters | |
| US20110274289A1 (en) | Sensor array beamformer post-processor | |
| JP4973657B2 (en) | Adaptive array control device, method, program, and adaptive array processing device, method, program | |
| CN103984676A (en) | Rectangular projection adaptive beamforming method based on covariance matrix reconstruction | |
| CN105137409B (en) | The sane space-time adaptive processing method of echo signal mutually constrained based on width | |
| CN113422630B (en) | Self-adaptive focusing broadband beam forming method and system | |
| CN111273237B (en) | Strong interference suppression method based on spatial matrix filtering and interference cancellation | |
| JPH10207490A (en) | Signal processor | |
| Zou et al. | A robust algorithm for linearly constrained adaptive beamforming | |
| Lovatello et al. | Steerable circular differential microphone arrays | |
| Kajala et al. | Filter-and-sum beamformer with adjustable filter characteristics | |
| CN111241470B (en) | Beam synthesis method and device based on self-adaptive null widening algorithm | |
| Fiori | Neural minor component analysis approach to robust constrained beamforming | |
| CN109669172B (en) | Weak target direction estimation method based on strong interference suppression in main lobe | |
| Zhang et al. | Selective frequency invariant uniform circular broadband beamformer | |
| Zhao et al. | On a particular family of differential beamformers with cardioid-like and no-null patterns | |
| Markovich‐Golan et al. | Spatial filtering | |
| CN113541650B (en) | Sparse linear constraint recursive maximum correlation entropy adaptive filter | |
| CN114755628A (en) | Method for estimating direction of arrival of acoustic vector sensor array under non-uniform noise | |
| US20220030353A1 (en) | Flexible differential microphone arrays with fractional order | |
| Zhang et al. | Frequency-invariant beamformer design via ADPM approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |