CN101419090B - Array noise signal focusing method in measurement of target noise - Google Patents

Array noise signal focusing method in measurement of target noise Download PDF

Info

Publication number
CN101419090B
CN101419090B CN2007101761512A CN200710176151A CN101419090B CN 101419090 B CN101419090 B CN 101419090B CN 2007101761512 A CN2007101761512 A CN 2007101761512A CN 200710176151 A CN200710176151 A CN 200710176151A CN 101419090 B CN101419090 B CN 101419090B
Authority
CN
China
Prior art keywords
noise signal
spectral line
noise
array
frequency
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.)
Expired - Fee Related
Application number
CN2007101761512A
Other languages
Chinese (zh)
Other versions
CN101419090A (en
Inventor
陈守虎
吴国清
张双荣
马力
曹建国
朱墨
周建清
肖迪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Acoustics CAS
Original Assignee
Institute of Acoustics CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Institute of Acoustics CAS filed Critical Institute of Acoustics CAS
Priority to CN2007101761512A priority Critical patent/CN101419090B/en
Publication of CN101419090A publication Critical patent/CN101419090A/en
Application granted granted Critical
Publication of CN101419090B publication Critical patent/CN101419090B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention relates to a method for focusing array noise signals during the process of target noise measurement. The method comprises the following steps: firstly, noise data of various channels is acquired; secondly, initial time-delay alignment is performed; thirdly, the noise data is analyzed and typical spectrum lines of the various channels are acquired; fourthly, phase information of the typical spectrum lines of the various channels are extracted; fifthly, refined phase alignment is performed; and sixthly, focusing signal is formed and output, namely focusing output of sound source targets can be obtained by superposition of various channel signals which are subjected to refined phase alignment. The method fully utilizes the phase information carried by narrowband compositions in target noise to solve the problem of insufficient focusing precision when only the prior distance measuring method is used for array focusing and realizes precise focusing of array measurement.

Description

The focus method of the array noise signal during a kind of target noise is measured
Technical field
The present invention relates to target noise and measure, the focus method of the array noise signal during particularly a kind of target noise is measured.
Background technology
Compare with common sound detection, the target noise measurement has following characteristics:
At first, the target source class may be very low, therefore, need measure by array usually, to obtain enough spatial gains;
Secondly, because the target source class is lower, simultaneously, need the radio-frequency component in the measuring-signal again, therefore, must carry out near field measurement to target, common measuring distance is 50～100m, on this distance, target radiated noise can not be considered as plane wave fully;
Once more, measured condition restriction, the shape of array, the position of array element all are not exclusively to determine.
In a word, because above-mentioned these factors, the precision that conventional wave beam forms is difficult to satisfactory.And the target noise measurement is not a qualitative detection, but measuring fixed amount requires to have higher measuring accuracy.
Based on above characteristics, the distance measuring signal that needs high s/n ratio when using array to carry out the target noise measurement is accurately measured the distance of target to each array element, the temporal information that provides by range-measuring circuit, can carry out delay operation to the signal of respective channel, this is that the array noise measuring system focuses on the typical practice of handling.Synchronous range finding is to measure one of reliable method of move distance between the cooperative target and measuring body under water at present, as its name suggests, range-measuring circuit is used to measure voice signal is arrived respective sensor by target propagation distance, in fact, what range-measuring circuit directly provided is the travel-time of voice signal by target arrival respective sensor, and multiply by the velocity of sound is exactly propagation distance.
But because the existence of factors such as signal distortion, device consistance, target volume unit, traditional distance accuracy often can only reach tens centimetres magnitude.And, for the consideration of system complexity and cost, can not be all array elements outfit range finding means usually, but on the basis of a few drive test distances, utilize the formation information geometry to find the solution other array element range-to-go.Because the incomplete determinacy of formation, the result of geometrical calculation will be introduced big range error to meter level.
For common ship noise signal, the range error of 30cm (delay time error that is equivalent to 0.2ms) will cause the signal correction performance to drop to 0.6, the range error of 1m (delay time error that is equivalent to 0.67ms) will cause the signal correction performance to drop to 0.4, such range error not only causes high-frequency signal focusing to fall flat, and can make the spatial manipulation gain of the low-frequency component loss 1～4dB of signal.So, be necessary how research under existing measuring condition, further improves the focusing accuracy of array noise signal.
Summary of the invention
The objective of the invention is in order to overcome the limitation of requirement that prior art does not reach the focusing accuracy of array noise signal, thereby the focus method of the array noise signal of a kind of target noise in measuring is provided, so that vernier focusing becomes possibility.
For realizing above-mentioned purpose of the present invention, core of the present invention is on the basis of routine range finding, the phase information of utilizing the low frequency spectral line to array element time-delay finely tune, thereby realize that accurate array focuses on.Because often there is the spectral line of some concentration of energy in the target noise spectrum, these spectral lines may be corresponding to engine speed, propeller blade number etc., simultaneously, because the generation mechanism of target noise, also because underwater acoustic channel is relatively stable in the close-in measurement, these spectral lines relatively steadily, are gradual in time accurate stationary signals at least often, and phase continuity is better.In addition, these spectral lines have often been concentrated stronger energy, the signal to noise ratio (S/N ratio) height, and the robustness of processing is better.
In signal analysis, usually the time resolution of narrow band signal that it is believed that spectral line one class is relatively poor, the reason that causes this received view is that narrow band signal has tangible short period feature, when using narrow band signal to carry out time resolution (being correlated with), separate (peak period corresponding to correlated results occurs) that be not unique such as using waveform.But in the one-period of narrow band signal, it is unique separating, and the restriction of one-period is equivalent to define a principal value interval to the problem of separating more.And existing range finding means can be locked in the sound path error rice or sub-meter grade, and for the low frequency signal below the 1kHz, on the basis of this range error, utilizing narrow band signal to carry out time resolution is exactly a problem that unique solution is arranged." preliminary phase alignment " stage is still continued to use this way among the present invention, so that the delay time error of different passage acoustical signals is controlled within the specific limits, becomes possibility thereby make next step use spectral line phase information to carry out vernier focusing.
The focus method of the array noise signal during a kind of target noise provided by the invention is measured may further comprise the steps:
(1) obtains the noise signal of each passage;
The target radiated noise that sensor array receives forms each channel noise signal through signals collecting recording apparatus collecting, record;
(2) preliminary time-delay alignment;
Ranging data by range-measuring circuit is to the alignment of tentatively delaying time of the noise signal of each passage;
(3) analyze noise signal, obtain the typical spectral line of each passage;
Calculate its third-octave average power spectral density by the power spectrum density of noise signal, the relatively power spectrum density and the third-octave average power spectral density of noise signal are chosen and are exceeded the above spectral line of third-octave average power spectral density 6dB as typical spectral line; Power spectrum by noise signal can calculate its third-octave average power spectra, promptly calculates the average power spectra in the third-octave frequency band, belongs to typical ripe algorithm, can be with reference to any books about power spectrumanalysis.
And the frequency of described typical spectral line should be lower than f u,
Figure 664608DEST_PATH_GSB00000281692200011
Wherein c is the velocity of sound, and Δ l is the distance accuracy of range-measuring circuit;
(4) extract the phase information of each passage typical case spectral line;
J channel noise burst is made fast fourier transform, R (f)+iI (f)=fft[X j(n)], according to the phase place of the corresponding frequencies position of the spectrum analysis j channel noise signal that filters out above, θ j ( f 0 ) = tan - 1 [ I ( f 0 ) R ( f 0 ) ] + π 2 { 1 - sgn [ R ( f 0 ) ] } ;
In the formula, f is the frequency on the frequency field; R (f) is the real part of noise signal power spectral density function after Fourier transform; I (f) is the imaginary part of noise signal power spectral density function after Fourier transform; X j(n) be the time series of j channel noise signal, wherein n represents natural sequence; f 0Be the typical spectral line frequency that is filtered out according to step (3);
(5) meticulous phase alignment;
With the 1st passage is reference, and the j passage should further be delayed time,
Figure 689382DEST_PATH_GSB00000281692200013
Wherein, symbol [] -π～πExpression by ± 2n π obtain principal value interval (π, π] on value;
(6) form focus signal output;
To just can obtain focusing output through each channel signal stack of meticulous phase alignment to acoustic target.
As a kind of selection of the present invention, described signals collecting pen recorder is sound-track engraving apparatus or the computing machine that has the signals collecting record cell.
As a kind of improvement of the present invention, the power spectrum density of described noise signal realizes by spectrum analyzer or by the power spectrum algorithm computation.The power spectrum density of calculating noise signal can realize by the power spectrum algorithm (period map method, Bartlett method, Welch method etc.) of spectrum analyzer or various maturations.
As further improvement of the present invention, it is characterized in that when the power spectrum density of noise signal was realized by the power spectrum algorithm computation, described power spectrum algorithm was period map method, Bartlett method or Welch method;
When the power spectrum algorithm was period map method, described period map method was,
Figure DEST_PATH_GSB00000204324400031
Wherein, w represents angular frequency, i.e. 2 π f; X (n) is the time series of noise signal, and N is the noise signal sequence length of being analyzed, f sBe the noise signal sampling rate.
As another improvement of the present invention, in described frequency less than f uTypical spectral line in, further select the high as far as possible spectral line of frequency as typical spectral line, or select the higher spectral line of power in the approaching spectral line of frequency as typical spectral line.
Rule of thumb, choose spectral line, it has been generally acknowledged that exceeding the above narrow band signal energy of third-octave average power spectra 6dB can regard as spectral line by the power spectrum and the third-octave average power spectra of comparison signal.But the spectral line that reaches this threshold value is understood more than one usually, selects to have three principles further:
The first, frequency should be lower than f u,
Figure DEST_PATH_GSB00000204324400032
Wherein c is the velocity of sound, and Δ l is the distance accuracy of range-measuring circuit.Frequency of utilization is higher than and equals f uThe phase information of spectral line focus on and can cause previously described many-valued uncertainty.
The second, less than f uSpectral line in, should select the high as far as possible spectral line of frequency, because the signal of relative high frequency rate has better time resolution precision.
The 3rd, for the approaching spectral line of frequency, should select the higher spectral line of power because such spectral line often corresponding certain basic mechanism of noise generation, signal to noise ratio (S/N ratio) is higher and relatively stable.
The invention has the advantages that, the focus method of the array noise signal during target noise of the present invention is measured, make full use of composition entrained phase information in arrowband in the target noise, solve the traditional distance measuring method of simple use and carried out the not enough difficulty of array focusing accuracy, realized the accurate focusing of array measurement.
Description of drawings
Fig. 1 is the schematic flow sheet of the focus method of the array noise signal during target noise of the present invention is measured.
Fig. 2 is that the system of one embodiment of the invention connects schematic block diagram.
Fig. 3 is the comparison diagram that forward and backward array signal correlated performance is finely tuned in the focusing of one embodiment of the invention.
Fig. 4 a and Fig. 4 b are respectively the synoptic diagram that forward and backward array signal waveform position relation is finely tuned in the focusing of one embodiment of the invention.
Embodiment
Be elaborated below in conjunction with the focus method of accompanying drawing to array noise signal of the present invention.
Present embodiment uses one section real ship noise data to carry out the laboratory simulation test of the focus method of array noise signal.As shown in Figure 2, digital signal-analog signal conversion the equipment controlled that uses a computer replaces the green end simulation to produce the received signal of nautical receiving set, under the export target motion measurement state, the ship noise signal that three-dimensional 16 nautical receiving sets that distribute receive, the original single channel sample of simulating signal is one section real ship noise.According to imaginary target trajectory and corresponding geometric relationship, stack 60kHz pure-tone polse is as distance measuring signal in noise signal, and dry end uses actual noise measuring equipment to carry out the signals collecting record.
As shown in Figure 1, in the present embodiment, the concrete steps of this focus method are as follows:
1. obtain each channel noise data
Target radiated noise is received by sensor array, gathers, writes down formation through signals collecting pen recorder (for example sound-track engraving apparatus).
2. use the ranging data alignment of tentatively delaying time
As its name suggests, range-measuring circuit is used to measure the propagation distance of voice signal by target arrival respective sensor, and in fact, what range-measuring circuit directly provided is the travel-time of voice signal by target arrival respective sensor, and multiply by the velocity of sound is exactly propagation distance.The temporal information of using range-measuring circuit to provide can be carried out delay operation to the signal of respective channel, and this also is that the array noise measuring system focuses on the typical practice of handling." preliminary phase alignment " stage is still continued to use this way among the present invention, so that the delay time error of different passage acoustical signals is controlled within the specific limits, becomes possibility thereby make next step use spectral line phase information to carry out vernier focusing.
3. the analysis noise data is obtained typical spectral line
This step operation can be divided into several sub-steps:
1) power spectrum density of calculating noise signal
Can realize by the power spectrum algorithm (period map method, Bartlett method, Welch method etc.) of spectrum analyzer or various maturations.With the period map method is example,
PSD ( w ) = 2 { abs ( fft [ X ( n ) ] ) } 2 N f s ,
Wherein X (n) is the time series of signal, and N is the burst length of being analyzed, f sBe signal sampling rate.
2) the third-octave average power spectra of calculating noise signal
Power spectrum by signal can calculate its third-octave average power spectra, promptly calculates the average power spectra in the third-octave frequency band.Owing to be typical ripe algorithm, can repeat no more here with reference to any books about power spectrumanalysis.
3) screening spectral line
Power spectrum and third-octave average power spectra by comparison signal are chosen spectral line.Rule of thumb, it has been generally acknowledged that exceeding the above narrow band signal energy of third-octave average power spectra 6dB can regard as spectral line.The spectral line that reaches this threshold value is understood more than one usually, selects to have three principles further:
The first, frequency should be lower than f u,
f u = c 4 Δl ,
Wherein c is the velocity of sound, and Δ l is the distance accuracy of range-measuring circuit.
Frequency of utilization is higher than and equals f uThe phase information of spectral line focus on and can cause previously described many-valued uncertainty.
The second, less than f uSpectral line in, should select the high as far as possible spectral line of frequency, because the signal of relative high frequency rate has better time resolution precision.
The 3rd, for the approaching spectral line of frequency, should select the higher spectral line of power because such spectral line often corresponding certain basic mechanism of noise generation, signal to noise ratio (S/N ratio) is higher and relatively stable.
4. extract the phase information of each passage typical case spectral line
With the j passage is example, and this channel signal sequence is made fast fourier transform,
R(f)+iI(f)＝fft[X j(n)]，
According to the phase place of the spectrum analysis corresponding frequencies position that filters out above,
θ j ( f 0 ) = tan - 1 [ I ( f 0 ) R ( f 0 ) ] + π 2 { 1 - sgn [ R ( f 0 ) ] } ,
5. meticulous phase alignment
If with the 1st passage is reference, the j passage should further be delayed time,
Δt = [ θ 1 ( f 0 ) - θ j ( f 0 ) ] - π ~ π 2 π f 0 ,
Wherein, symbol [] -π～πExpression is by ± 2n π, obtain principal value interval (π, π] on value.
6. form focus signal output
To just can obtain focusing output through each channel signal stack of meticulous phase alignment to acoustic target.
Test findings shows, directly output focuses on the correlated results (is reference channel with 16 passages) of each array element signals of time-delay back shown in the curve that asterisk among Fig. 3 indicates according to range-measuring circuit, get relevant relatively poor passage 9 signals and observe waveform with reference channel, shown in Fig. 4 a, signal sampling rate is 50kHz among the figure.As can be seen from the figure, directly use the general skew that has 10 sampled points (corresponding range error is 30cm) of result of original ranging data time-delay.The analytic signal power spectrum, there is a stronger spectral line in discovery at the 123Hz place, after using the phase information of this spectral line that each array element signals is carried out fine delay, signal waveform is shown in Fig. 4 b, and the correlated results of each array element signals is shown in the curve that plus sige among Fig. 3 indicates.As seen, after fine setting, the correlated performance of signal has obtained obvious improvement.

Claims (6)

1. the focus method of the array noise signal during a target noise is measured may further comprise the steps:
(1) obtains the noise signal of each passage;
The target radiated noise that sensor array receives forms each channel noise signal through signals collecting recording apparatus collecting, record;
(2) preliminary time-delay alignment;
Ranging data by range-measuring circuit is to the alignment of tentatively delaying time of the noise signal of each passage;
(3) analyze noise signal, obtain the typical spectral line of each passage;
Calculate its third-octave average power spectral density by the power spectrum density of noise signal, the relatively power spectrum density and the third-octave average power spectral density of noise signal are chosen and are exceeded the above spectral line of third-octave average power spectral density 6dB as typical spectral line; The frequency of described typical spectral line should be lower than f u,
Figure RE-FSB00000281692100011
Wherein c is the velocity of sound, and Δ l is the distance accuracy of range-measuring circuit;
(4) extract the phase information of each passage typical case spectral line;
J channel noise burst is made fast fourier transform, R (f)+iI (f)=fft[X j(n)], the phase place of the corresponding frequencies position of the spectrum analysis j channel noise signal that filters out according to step (3),
In the formula, f is the frequency on the frequency field; R (f) is the real part of noise signal power spectral density function after Fourier transform; I (f) is the imaginary part of noise signal power spectral density function after Fourier transform; X j(n) be the time series of j channel noise signal, wherein n represents natural sequence; f 0Be the typical spectral line frequency that is filtered out according to step (3);
(5) meticulous phase alignment;
With the 1st passage is reference, and the j passage should further be delayed time,
Figure RE-FSB00000281692100013
Wherein, symbol [] -π-πExpression by ± 2n π obtain principal value interval (π, π] on value;
(6) form focus signal output;
To just can obtain focusing output through each channel noise signal stack of meticulous phase alignment to acoustic target.
2. the focus method of array noise signal according to claim 1 is characterized in that, described signals collecting pen recorder is sound-track engraving apparatus or the computing machine that has the signals collecting record cell.
3. the focus method of array noise signal according to claim 1 is characterized in that, the power spectrum density of described noise signal realizes by spectrum analyzer or by the power spectrum algorithm computation.
4. the focus method of array noise signal according to claim 3 is characterized in that, when the power spectrum density of noise signal was realized by the power spectrum algorithm computation, described power spectrum algorithm was period map method, Bartlett method or Welch method.
5. the focus method of array noise signal according to claim 4 is characterized in that, when the power spectrum algorithm was period map method, described period map method was:
Figure FSB00000204324300021
Wherein, w represents angular frequency, i.e. 2 π f; X (n) is the time series of noise signal, and N is the noise signal sequence length of being analyzed, f sBe the noise signal sampling rate.
6. the focus method of array noise signal according to claim 1 is characterized in that, the frequency that filters out in step (3) is less than f uTypical spectral line in, further select the high as far as possible spectral line of frequency as typical spectral line, or select the higher spectral line of power in the approaching spectral line of frequency as typical spectral line.
CN2007101761512A 2007-10-22 2007-10-22 Array noise signal focusing method in measurement of target noise Expired - Fee Related CN101419090B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2007101761512A CN101419090B (en) 2007-10-22 2007-10-22 Array noise signal focusing method in measurement of target noise

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2007101761512A CN101419090B (en) 2007-10-22 2007-10-22 Array noise signal focusing method in measurement of target noise

Publications (2)

Publication Number Publication Date
CN101419090A CN101419090A (en) 2009-04-29
CN101419090B true CN101419090B (en) 2011-03-16

Family

ID=40629991

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2007101761512A Expired - Fee Related CN101419090B (en) 2007-10-22 2007-10-22 Array noise signal focusing method in measurement of target noise

Country Status (1)

Country Link
CN (1) CN101419090B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101738611B (en) * 2009-12-15 2012-08-15 中国科学院声学研究所 Underwater acoustic target signal detection and identification method
US8625902B2 (en) * 2010-07-30 2014-01-07 Qualcomm Incorporated Object recognition using incremental feature extraction
CN102692314B (en) * 2011-03-22 2014-12-03 浙江大学 Apparatus and method for testing power spectral density of frequency noise of laser based on fiber resonator
CN111175727B (en) * 2018-11-13 2022-05-03 中国科学院声学研究所 Method for estimating orientation of broadband signal based on conditional wave number spectral density
CN114088332B (en) * 2021-11-24 2023-08-22 成都流体动力创新中心 Wind tunnel background noise correction method for rotating blade sound signal extraction

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
吴国清等.用垂直阵和单水听器测量水下目标辐射噪声的误差分析及其修正方法.声学学报32 5.2007,32(5),398-403.
吴国清等.用垂直阵和单水听器测量水下目标辐射噪声的误差分析及其修正方法.声学学报32 5.2007,32(5),398-403. *
陈守虎等.噪声测量中利用目标谱线进行精确波束形成的研究.中国声学学会2006年全国声学学术会议论文集.2006,(2006),47-48. *

Also Published As

Publication number Publication date
CN101419090A (en) 2009-04-29

Similar Documents

Publication Publication Date Title
CN101470187B (en) High-precision direction finding method used for linear array
CN101098179B (en) Method for forming broadband frequency domain digital beam
CN101419090B (en) Array noise signal focusing method in measurement of target noise
CN101644773B (en) Real-time frequency domain super-resolution direction estimation method and device
CN101702000A (en) Phased ultrasound locating method and system of partial discharge of transformer
CN108509377B (en) Pulse signal arrival time and pulse width estimation method based on edge feature extraction
CN104678384B (en) Method for estimating underwater target speed by using sound pressure difference cross-correlation spectrum analysis of beam fields
CN103063909B (en) A kind of linear frequency-modulated parameter estimation method based on power spectrum
CN103293515A (en) Ship and warship line spectrum noise source longitudinal distribution characteristic measuring method
CN109991590B (en) System and method for testing low-frequency emission characteristic of transducer in pressure tank in limited space
CN109001297B (en) Method for measuring acoustic reflection coefficient of large-sample underwater acoustic material based on single-vector hydrophone
CN106646350A (en) Correction method for inconsistency of amplitude gains of channels of single vector hydrophone
CN112859019A (en) Intra-pulse modulation type parameter extraction system and using method
CN102495280B (en) Anti-noise wide-range frequency measurement method and phase locking frequency meter
CN110346752A (en) Nothing based on relatively prime Sparse Array obscures direction-finding method
CN105158751A (en) Acoustic vector array fast DOA (Direction of Arrival) estimation method
CN103645248A (en) High-temperature alloy grain size evaluation method based on ultrasonic phase velocity
CN102540153A (en) Array amplitude and phase error correcting method based on interstation direct wave interference
CN105738482A (en) Ultrasonic phased array beam delay method for FIR (Finite Impulse Response) multi-phase interpolation filtering
CN112230212B (en) Radar ranging signal processing method and device
CN106330342A (en) Water sound communication doppler factor estimation method with low computation complexity
CN107783135A (en) A kind of three-element vector battle array passive ranging method
CN110703186A (en) Bi-hydrophone phase difference calibration and measurement system and method for near field of pool
CN116819432A (en) Single-vector hydrophone underwater multi-target high-stability direction finding method and system based on characteristic spectrum tracking
CN105745550A (en) Analog to information converter

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20110316

Termination date: 20131022