CN109238447A - A kind of blind source separation method of tether vibration signal - Google Patents
A kind of blind source separation method of tether vibration signal Download PDFInfo
- Publication number
- CN109238447A CN109238447A CN201811062499.3A CN201811062499A CN109238447A CN 109238447 A CN109238447 A CN 109238447A CN 201811062499 A CN201811062499 A CN 201811062499A CN 109238447 A CN109238447 A CN 109238447A
- Authority
- CN
- China
- Prior art keywords
- signal
- vibration
- matrix
- tether
- blind source
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M7/00—Vibration-testing of structures; Shock-testing of structures
- G01M7/02—Vibration-testing by means of a shake table
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
A kind of blind source separation method of tether vibration signal, comprising the following steps: Step 1: selecting any position point on tether obtains vibration signal；Step 2: carrying out mean value pretreatment to collected vibration signal, go mean value that observation signal is made to become zero mean vector；Step 3: removing the correlation between observation signal to going the observation signal data after mean value to carry out whitening processing；Step 4: the observation signal after whitening processing is analyzed to obtain the estimation of hybrid matrix and source signal by blind source separation algorithm；Step 5: analyzing the modal response and vibration shape matrix of tether vibrational system according to the estimation of hybrid matrix and source signal；Using single mode identification technology from source signal estimate in isolate frequency and damping ratio；The each signal isolated is sized to get frequency vector, damping ratio vector and the vibration shape matrix of tether vibrational system is arrived.System parameter recognition accuracy of the invention is higher.
Description
Technical field
The present invention relates to tether vibrational system analysis methods, and in particular to a kind of blind source separating side of tether vibration signal
Method.
Background technique
In life and production usually, tether arrangements are widely present, such as the laying of power transmission line, the processing of braided wire,
Elevator lifting system and the design of cable-stayed bridge etc..In tether arrangements application, tether vibration is a kind of generally existing phenomenon,
Vibration signal includes many parameters system-related, such as tension, and tension is directly related to the reliability of power transmission line, braided wire
The safety of quality and elevator and cable-stayed bridge.It obtains tether vibration signal and analyzes it, to obtain desired parameter.
The vibration signal obtained by sensor (contact and contactless and deposit) is mostly clutter, comprising noise and
Useless ingredient, and the ratio of ingredient is unknown.Therefore, how the source signal of accurate analysis system, be to solve for many years
Certainly the problem of.Commonly used two methods, emphasis is respectively with sensor and signal sheet.For sensor, adopt nothing more than
Accuracy when acquiring is improved with new technology；For signal itself, the extraction of useful information is got down to, the rejecting with interference.Mesh
Front signal processing method include IIR filter method, FIR filter method, wavelet theory method, statistic line loss rate method, Time-frequency Analysis,
Self-adaptive routing, neural network signal processing method etc., the above method are to divide under known transmission channel signal
Analysis.This is entirely different with blind source separating method, and blind source separating method is situation about all can not accurately determine in source signal and transmission channel
Under, only analysis obtains the signal processing method of expectation parameter value from collected clutter.The method is applied to be believed in vibration
Number analysis when, agree with very much with modal analysis method.For vibrational system, modal analysis method is that the every first order mode of system is opposite
Modal coordinate, which marks off, to be come, and the differential equation coupled under physical coordinates is obtained, independent under modal coordinate, to represent and be
The parameter value of every rank mode in system.Blind source separating method is signal processing method, and modal analysis method is common in Structural Dynamics
Method, both of which are the emphasis that will it is expected parameter " independence " as research, carry out Separation Research with this.
It is mostly the measurement using contact in the acquisition modes of tether parameter, the mode of this measurement exists to structure
Interference, with the development of sensor technology, contactless measurement method have great development prospect and research significance.It is non-
The measurement method of contact can choose high speed camera, linear CCD sensor, PSD position sensor and laser mouse chip
Deng.In the selection of non-cpntact measurement, there are the difference of overall situation and partial situation, the data volume of global measuring is big, and equipment requirement is high, accurately
Degree also can highest.But in the selection of actual measurement, in global method no longer limit of consideration, because the measurement of part can expire
The demand of sufficient signal acquisition.The core of non-cpntact measurement is the research to tether vibration motion, and current analysis method has difference
Divide method, standing wave method and frequency method.String is in vibration processes, the damping of tether, the interference of frictional force, can all vibrate to tether
System, which is constituted, to be influenced, and is restricted similar to the output channel of signal, can not accurately be analyzed its variation.
Summary of the invention
It is an object of the invention to be directed to above-mentioned the problems of the prior art, a kind of blind source point of tether vibration signal is provided
From method, the vibration shape and modal displacement when tether vibrates are researched and analysed from the displacement time curve of multiple points, accurately obtains system
Parameter.
To achieve the goals above, the technical solution adopted by the present invention the following steps are included:
Step 1: selecting any position point on tether obtains vibration signal；
Step 2: to collected vibration signal carry out mean value pretreatment, go mean value make observation signal become zero-mean to
Amount；
Step 3: removing the correlation between observation signal to going the observation signal data after mean value to carry out whitening processing；
Step 4: observation signal after whitening processing analyzes to obtain hybrid matrix and source signal by blind source separation algorithm
Estimation；
Step 5: analyzing the modal response and vibration shape square of tether vibrational system according to the estimation of hybrid matrix and source signal
Battle array；
Using single mode identification technology from source signal estimate in isolate frequency and damping ratio；To each signal isolated by
Frequency vector, damping ratio vector and the vibration shape matrix of tether vibrational system are arrived according to big minispread.
For step 1 when obtaining vibration signal, the expression formula of vibration response signal is as follows:
In formula, ωnjFor system frequency, ωdjFor the intrinsic frequency in the presence of damping, ΦjFor complex mode vibration shape vector,
ΦijFor vibration shape vector the point complex values,For ΦijConjugate complex number, qjFor response of mode displacement,For qjConjugate complex
Number, σijFor phase angle, ξjFor damping ratio；gijTo assume coefficient, characteristic value sjWith vibration shape vector ΦjIt is the conjugation occurred in pairs
Complex value, αjFor real number.
Observation signal indicates are as follows:
Xbss(t)=AS (t)+n (t)
In formula, n (t) indicates mixed noise signal in signal acquisition process；By blind source separation algorithm from observation signal
In isolate the hybrid matrix A and source signal s (t) of system, then extrapolate separation matrix W.
Step 3 carries out whitening processing to the observation signal data after going mean value, and specific step is as follows:
A. the covariance matrix of observation signal is indicated are as follows: Rx=E [XXT]；X in formula is mixed to be obtained by observation signal
Close matrix signal, RxFor the covariance matrix of hybrid matrix signal X, E is mathematic expectaion；
B. Eigenvalues Decomposition is carried out as the following formula to the covariance matrix of observation signal:
V in formulaxIt is characterized vector matrix, DxDiagonal values for diagonal matrix, diagonal matrix are characterized value；
C. the whitening matrix is enabled to beWhitened signal is expressed asAfter Data Whitening
Unit covariance matrix isI in formula is unit battle array,For whitened signal.
Blind source separation algorithm described in step 4 selects FastICA, SNRMax, AMUSE or SOBI algorithm.
Step 5 in actual analysis hybrid matrix A, source signal s (t) and vibration shape vector Φ, modal response q (t) and point
From matrix W, there are following relationships: Φ ≈ A=W-1；q(t)≈s(t).
Single mode identification technology described in step 5 includes time domain peak fitting process and half power bandwidth method.
Compared with prior art, the present invention have it is following the utility model has the advantages that the blind source separation method on tether to position
Point it is selected there is no limit, acquire be entire tether vibrational system frequency vector, damping ratio vector and vibration shape matrix, and
And because the vibration of tether is mainly low order vibration, each measurement point corresponds to single order vibration, so the quantity of location point can
Four are set as, due to being analyzed with the vibration of single point, interference when to signal acquisition has great inhibiting effect.Between
The characteristic of blind source separation method, i.e., to transmission channel without limitation, so the placement of measurement point can follow structure and design optimal original
Then.Mean value and whitening pretreatment are carried out to the data of acquisition first, reduce the difficulty of subsequent step processing.Then to obtained number
Its source signal and hybrid matrix are analyzed according to being brought into blind source separation algorithm, correspondence obtains the modal response and vibration of system
Type.Finally using the methods of single mode identification technology, the parameter in vibrational system is identified.Parameter recognition accuracy of the invention
It is higher.
Detailed description of the invention
The placement schematic diagram of Fig. 1 measurement point of the present invention；
The variation schematic diagram of Fig. 2 vibration signal of the present invention.
Specific embodiment
Present invention will be described in further detail below with reference to the accompanying drawings.
The blind source separation method of tether vibration signal of the present invention, comprising the following steps:
One, the acquisition of vibration signal vibrates (small damping) system for tether, and vibration response signal can be expressed as follows:
In formula, ωnjFor system frequency, ωdjFor the intrinsic frequency in the presence of damping, ΦjFor complex mode vibration shape vector,
ΦijFor vibration shape vector the point complex values,For ΦijConjugate complex number, qjFor response of mode displacement,For qjConjugation
Plural number, σijFor phase angle, ξjFor damping ratio；gijTo assume coefficient, characteristic value sjWith vibration shape vector ΦjIt is being total to of occurring in pairs
Yoke complex value, αjFor real number.
The mode of oscillation of tether, it is larger with low order mode proportion, so the number selection of measurement point can at four
It meets the requirements.As shown in Figure 1, selecting any position point A, B, C, D on tether obtains vibration signal, as wanting signal
The input of processing, by blind source separation method, it is found that it is to signal input channel, there is no limitations.
Two, mean value pretreatment is carried out to the signal of acquisition, under normal circumstances, observation signal can be expressed as follows:
Xbss(t)=AS (t) (3)+n (t)
In formula (3), n (t) indicates mixed noise signal in signal acquisition process, by taking zero mean Gaussian white noise as an example.
As shown in Fig. 2, blind source separating is exactly that the hybrid matrix A for isolating system from the clutter observed and source are believed
Number s (t), so separation matrix W is the main pursuit object of separation task.Mean value is gone to observation signal in blind source separating (BBS),
That is the mean value K=E { x } of observation signal subtraction signal, makes observation signal become zero mean vector.The source signal s obtained with BBS
(t) estimation Y is also zero-mean.This step pretreatment can simplify the operation of BBS, inessential condition.Data after spending mean value
After extrapolating separation matrix W, the mean value vector removed is added to be the estimation approximation of source signal s (t) on source signal estimation Y,
The mean value vector removed is W × K, and K is the mean value that mean value is removed in the process.
Three, after going mean value to observation data, whitening processing is carried out, in order that the correlation between removal observation signal,
Simplify subsequent separation algorithm, improves the stability of algorithm.Signal after albefaction is known as whitened signal, under normal circumstances, data
Albefaction be to be completed by the method for principal component analysis, detailed process is described below:
The covariance matrix of observation signal first, may be expressed as:
Rx=E [XXT] (4)
Wherein: X is the hybrid matrix signal obtained by observation signal, RxFor the covariance matrix of hybrid matrix signal X, E
For mathematic expectaion.Eigenvalues Decomposition is carried out to formula (4), as follows:
Wherein, VxIt is characterized vector matrix, DxFor diagonal matrix (diagonal values are characterized value).
Enable whitening matrix are as follows:
Whitened signal may be expressed as:
It is unit covariance matrix by formula (7) it is found that after Data Whitening, it may be assumed that
Wherein, I is unit battle array,For whitened signal.
When observation signal number is greater than the rank number of mode of system, whitening pretreatment is carried out to observation signal, it can be effective
Determination system mode order, pretreatment equally also have beneficial effect to the filtering of white noise.
Four, the observation signal after whiteningUsing blind source separation algorithm (such as FastICA, SNRMax, AMUSE, SOBI)
Analysis, can be obtained the estimation of hybrid matrix A and source signal s (t).
The core of blind source separation algorithm is to analyze the correlation in signal, can be indicated with different means, such as related
Property, entropy or gauss of distribution function value etc. are the processes of an iteration, until meeting precision.
Five, the source signal s (t) and hybrid matrix A obtained according to algorithm estimates, analyzes the modal response of tether vibrational system
Q (t) (including frequency, damping ratio) and vibration shape matrix Φ.In actual analysis, hybrid matrix A, source signal s (t) and vibration shape vector
There are following relationships by Φ, modal response q (t) and separation matrix W:
Φ ≈ A=W-1 (9)
q(t)≈s(t) (10)
Divide from source signal estimation s (t) using single mode identification technology (such as time domain peak fitting process, half power bandwidth method etc.)
Separate out frequency and damping ratio.Because the signal sequence that blind source sub-argument goes out is indefinite, each signal isolated is sized,
The frequency vector, damping ratio vector and vibration shape matrix of tether vibrational system can be obtained.
Claims (7)
1. a kind of blind source separation method of tether vibration signal, which comprises the following steps:
Step 1: selecting any position point on tether obtains vibration signal；
Step 2: carrying out mean value pretreatment to collected vibration signal, go mean value that observation signal is made to become zero mean vector；
Step 3: removing the correlation between observation signal to going the observation signal data after mean value to carry out whitening processing；
Step 4: observation signal after whitening processing analyzes to obtain estimating for hybrid matrix and source signal by blind source separation algorithm
Meter；
Step 5: analyzing the modal response and vibration shape matrix of tether vibrational system according to the estimation of hybrid matrix and source signal；
Using single mode identification technology from source signal estimate in isolate frequency and damping ratio；To each signal isolated according to big
Minispread is to get frequency vector, damping ratio vector and the vibration shape matrix for arriving tether vibrational system.
2. the blind source separation method of tether vibration signal according to claim 1, which is characterized in that step 1 is obtaining vibration
When signal, the expression formula of vibration response signal is as follows:
In formula, ωnjFor system frequency, ωdjFor the intrinsic frequency in the presence of damping, ΦjFor complex mode vibration shape vector, Φij
For vibration shape vector the point complex values,For ΦijConjugate complex number, qjFor response of mode displacement,For qjConjugate complex number,
σijFor phase angle, ξjFor damping ratio；gijTo assume coefficient, characteristic value sjWith vibration shape vector ΦjIt is the conjugate complex occurred in pairs
Value, αjFor real number.
3. the blind source separation method of tether vibration signal according to claim 1, which is characterized in that observation signal indicates are as follows:
Xbss(t)=AS (t)+n (t)
In formula, n (t) indicates mixed noise signal in signal acquisition process；Divided from observation signal by blind source separation algorithm
The hybrid matrix A and source signal s (t) for separating out system, then extrapolate separation matrix W.
4. the blind source separation method of tether vibration signal according to claim 1, which is characterized in that after step 3 is to mean value is gone
Observation signal data carry out whitening processing specific step is as follows:
A. the covariance matrix of observation signal is indicated are as follows: Rx=E [XXT]；X in formula is the mixed moment obtained by observation signal
Battle array signal, RxFor the covariance matrix of hybrid matrix signal X, E is mathematic expectaion；
B. Eigenvalues Decomposition is carried out as the following formula to the covariance matrix of observation signal:
V in formulaxIt is characterized vector matrix, DxDiagonal values for diagonal matrix, diagonal matrix are characterized value；
C. the whitening matrix is enabled to beWhitened signal is expressed asUnit after Data Whitening
Covariance matrix isI in formula is unit battle array,For whitened signal.
5. the blind source separation method of tether vibration signal according to claim 1, which is characterized in that blind source described in step 4
Separation algorithm selects FastICA, SNRMax, AMUSE or SOBI algorithm.
6. the blind source separation method of tether vibration signal according to claim 1, which is characterized in that step 5 is in actual analysis
There are following relationships by middle hybrid matrix A, source signal s (t) and vibration shape vector Φ, modal response q (t) and separation matrix W:
Φ ≈ A=W-1
q(t)≈s(t)。
7. the blind source separation method of tether vibration signal according to claim 1, which is characterized in that single mode described in step 5
State identification technology includes time domain peak fitting process and half power bandwidth method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811062499.3A CN109238447A (en) | 2018-09-12 | 2018-09-12 | A kind of blind source separation method of tether vibration signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811062499.3A CN109238447A (en) | 2018-09-12 | 2018-09-12 | A kind of blind source separation method of tether vibration signal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109238447A true CN109238447A (en) | 2019-01-18 |
Family
ID=65060889
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811062499.3A Pending CN109238447A (en) | 2018-09-12 | 2018-09-12 | A kind of blind source separation method of tether vibration signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109238447A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109992834A (en) * | 2019-03-05 | 2019-07-09 | 中国人民解放军海军勤务学院 | The distinguishing structural mode method of modified blind source separating |
| CN110782041A (en) * | 2019-10-18 | 2020-02-11 | 哈尔滨工业大学 | Structural modal parameter identification method based on machine learning |
| CN112565119A (en) * | 2020-11-30 | 2021-03-26 | 西北工业大学 | Broadband DOA estimation method based on time-varying mixed signal blind separation |
| CN113221986A (en) * | 2021-04-30 | 2021-08-06 | 西安理工大学 | Method for separating vibration signals of through-flow turbine |
| CN113671037A (en) * | 2021-09-01 | 2021-11-19 | 杭州意能电力技术有限公司 | Post insulator vibration acoustic signal processing method |
| CN114925726A (en) * | 2022-05-13 | 2022-08-19 | 华侨大学 | Method, device, equipment and storage medium for identifying sub-sampling working mode parameters |
| CN116866124A (en) * | 2023-07-13 | 2023-10-10 | 中国人民解放军战略支援部队航天工程大学 | Blind separation method based on baseband signal time structure |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102288285A (en) * | 2011-05-24 | 2011-12-21 | 南京航空航天大学 | Blind source separation method for single-channel vibration signals |
| US20140296726A1 (en) * | 2009-11-03 | 2014-10-02 | Vivaquant Llc | Method and Apparatus for Identifying Cardiac Risk |
| CN104674644A (en) * | 2015-02-05 | 2015-06-03 | 中铁大桥局集团武汉桥梁科学研究院有限公司 | Vibration control device and method for slings of suspension bridge |
| CN106970635A (en) * | 2017-05-17 | 2017-07-21 | 西北工业大学 | It is a kind of to pull the tether swing suppressing method for becoming in rail and utilizing towing aircraft motor-driven |
| CN107192448A (en) * | 2017-05-17 | 2017-09-22 | 苏交科集团股份有限公司 | Peak method is searched in a kind of broadband for recognizing flexible rope vibration frequency |
| CN107314807A (en) * | 2017-07-07 | 2017-11-03 | 河南科技大学 | A kind of method of steel wire rope horizontal space vibration measurement |
| CN107356419A (en) * | 2017-07-18 | 2017-11-17 | 厦门大学 | A kind of experimental method for being used to measure rope damping parameter |
| CN107357977A (en) * | 2017-06-27 | 2017-11-17 | 华侨大学 | The recognition methods of linear structure operational modal parameter and device based on second-order blind identification |
| CN206656855U (en) * | 2017-04-24 | 2017-11-21 | 苏交科集团股份有限公司 | A kind of device for laboratory measurement flexible cable frequency and the vibration shape |
| CN107633132A (en) * | 2017-09-18 | 2018-01-26 | 湖南大学 | The equivalent recognition methods of spatial distribution dynamic load |
-
2018
- 2018-09-12 CN CN201811062499.3A patent/CN109238447A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140296726A1 (en) * | 2009-11-03 | 2014-10-02 | Vivaquant Llc | Method and Apparatus for Identifying Cardiac Risk |
| CN102288285A (en) * | 2011-05-24 | 2011-12-21 | 南京航空航天大学 | Blind source separation method for single-channel vibration signals |
| CN104674644A (en) * | 2015-02-05 | 2015-06-03 | 中铁大桥局集团武汉桥梁科学研究院有限公司 | Vibration control device and method for slings of suspension bridge |
| CN206656855U (en) * | 2017-04-24 | 2017-11-21 | 苏交科集团股份有限公司 | A kind of device for laboratory measurement flexible cable frequency and the vibration shape |
| CN106970635A (en) * | 2017-05-17 | 2017-07-21 | 西北工业大学 | It is a kind of to pull the tether swing suppressing method for becoming in rail and utilizing towing aircraft motor-driven |
| CN107192448A (en) * | 2017-05-17 | 2017-09-22 | 苏交科集团股份有限公司 | Peak method is searched in a kind of broadband for recognizing flexible rope vibration frequency |
| CN107357977A (en) * | 2017-06-27 | 2017-11-17 | 华侨大学 | The recognition methods of linear structure operational modal parameter and device based on second-order blind identification |
| CN107314807A (en) * | 2017-07-07 | 2017-11-03 | 河南科技大学 | A kind of method of steel wire rope horizontal space vibration measurement |
| CN107356419A (en) * | 2017-07-18 | 2017-11-17 | 厦门大学 | A kind of experimental method for being used to measure rope damping parameter |
| CN107633132A (en) * | 2017-09-18 | 2018-01-26 | 湖南大学 | The equivalent recognition methods of spatial distribution dynamic load |
Non-Patent Citations (2)
| Title |
|---|
| 张晓丹: "《基于盲源分离技术的工程结构模态参数识别方法研究》", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 黄方林等: "《随机减量法在斜拉桥拉索模态参数识别中的应用》", 《机械强度》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109992834A (en) * | 2019-03-05 | 2019-07-09 | 中国人民解放军海军勤务学院 | The distinguishing structural mode method of modified blind source separating |
| CN110782041A (en) * | 2019-10-18 | 2020-02-11 | 哈尔滨工业大学 | Structural modal parameter identification method based on machine learning |
| CN110782041B (en) * | 2019-10-18 | 2022-08-02 | 哈尔滨工业大学 | Structural modal parameter identification method based on machine learning |
| CN112565119A (en) * | 2020-11-30 | 2021-03-26 | 西北工业大学 | Broadband DOA estimation method based on time-varying mixed signal blind separation |
| CN112565119B (en) * | 2020-11-30 | 2022-09-27 | 西北工业大学 | Broadband DOA estimation method based on time-varying mixed signal blind separation |
| CN113221986A (en) * | 2021-04-30 | 2021-08-06 | 西安理工大学 | Method for separating vibration signals of through-flow turbine |
| NL2030483A (en) * | 2021-04-30 | 2022-11-15 | Univ Xian Technology | Test System and Method for Separating Vibration Signals of Tubular Water Turbines |
| CN113671037A (en) * | 2021-09-01 | 2021-11-19 | 杭州意能电力技术有限公司 | Post insulator vibration acoustic signal processing method |
| CN114925726A (en) * | 2022-05-13 | 2022-08-19 | 华侨大学 | Method, device, equipment and storage medium for identifying sub-sampling working mode parameters |
| CN116866124A (en) * | 2023-07-13 | 2023-10-10 | 中国人民解放军战略支援部队航天工程大学 | Blind separation method based on baseband signal time structure |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109238447A (en) | A kind of blind source separation method of tether vibration signal | |
| CN103971124B (en) | Multi-class motor imagery brain electrical signal classification method based on phase synchronization | |
| CN108924736B (en) | PCA-Kalman-based passive indoor personnel state detection method | |
| CN105044590B (en) | A kind of single measuring node analog-circuit fault diagnosis method | |
| CN110837768B (en) | Online detection and identification method for rare animal protection | |
| CN109243483A (en) | A kind of noisy frequency domain convolution blind source separation method | |
| CN105654963B (en) | Voice underdetermined blind recognition method and device by spectrum correction and data density clustering method | |
| CN111241904B (en) | Operation mode identification method under underdetermined condition based on blind source separation technology | |
| US20150293143A1 (en) | Feature Extraction from Human Gaiting Patterns using Principal Component Analysis and Multivariate Empirical Mode Decomposition | |
| CN109583350A (en) | A kind of high-precision denoising method of local ultrasound array signal | |
| CN109884591A (en) | A kind of multi-rotor unmanned aerial vehicle acoustical signal Enhancement Method based on microphone array | |
| CN108491608A (en) | The Sparse Component Analysis method of distinguishing structural mode when number of sensors is incomplete | |
| CN111735523A (en) | Vehicle weight detection method and device based on video identification and storage medium | |
| CN106248201B (en) | Harmonic components detection method based on enhancing spectrum kurtosis | |
| CN112394324A (en) | Microphone array-based remote sound source positioning method and system | |
| CN103295193A (en) | Cross-power spectrum based blind source separation method | |
| CN110458071B (en) | DWT-DFPA-GBDT-based optical fiber vibration signal feature extraction and classification method | |
| CN110782041B (en) | Structural modal parameter identification method based on machine learning | |
| CN116343261A (en) | Gesture recognition method and system based on multi-modal feature fusion and small sample learning | |
| CN107941324A (en) | A kind of method of estimation of consumer level inertia sensing unit environments noise | |
| Virta et al. | Determining the signal dimension in second order source separation | |
| Yin et al. | Clustering distributed time series in sensor networks | |
| CN112163540B (en) | Gesture recognition method based on WiFi | |
| CN109711468A (en) | A kind of complex value blind separation hybrid matrix estimation method based on single source point | |
| CN115826042B (en) | Edge cloud combined distributed seismic data processing method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |
|
| RJ01 | Rejection of invention patent application after publication |