CN108599773A - A kind of vibration signal data compression acquisition method based on certainty calculation matrix - Google Patents
A kind of vibration signal data compression acquisition method based on certainty calculation matrix Download PDFInfo
- Publication number
- CN108599773A CN108599773A CN201810336121.1A CN201810336121A CN108599773A CN 108599773 A CN108599773 A CN 108599773A CN 201810336121 A CN201810336121 A CN 201810336121A CN 108599773 A CN108599773 A CN 108599773A
- Authority
- CN
- China
- Prior art keywords
- matrix
- calculation matrix
- certainty calculation
- orthogonal
- certainty
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/147—Discrete orthonormal transforms, e.g. discrete cosine transform, discrete sine transform, and variations therefrom, e.g. modified discrete cosine transform, integer transforms approximating the discrete cosine transform
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/12—Analogue/digital converters
- H03M1/124—Sampling or signal conditioning arrangements specially adapted for A/D converters
Abstract
The invention discloses a kind of, and the vibration signal data based on certainty calculation matrix compresses acquisition method, includes the following steps：The vibration signal for extracting acquisition carries out sparsity analysis under dct basis DCT orthogonal basis, obtains degree of rarefication K；According to degree of rarefication K, the first Symmetric Orthogonal certainty calculation matrix is established based on certainty calculation matrix；The first Symmetric Orthogonal certainty calculation matrix of acquisition is iterated using threshold value iterative shrinkage algorithm, obtains the second Symmetric Orthogonal certainty calculation matrix；The second obtained Symmetric Orthogonal certainty calculation matrix is optimized using singular value decomposition algorithm, obtains third Symmetric Orthogonal certainty calculation matrix.Vibration signal data compression acquisition method proposed by the present invention based on certainty calculation matrix is solved because the reconstruction error of certainty calculation matrix is larger, the problem of failing to be used widely.
Description
Technical field
The invention belongs to data compressions to acquire field, and in particular to a kind of vibration signal number based on certainty calculation matrix
According to compression acquisition method.
Background technology
The vibration signals collecting being widely adopted at present is adopted with the modulus (A/D) that Nyquist sampling theorem is guidance
Sample, but its sample frequency is not less than twice of original signal highest frequency, ability Exact Reconstruction goes out the signal.However, nowadays
Mechanical equipment high speed and enlargement increasingly, vibration frequency is higher and higher and non-linear, non-stationary feature is presented.If
It is still sampled with traditional sampling thheorem, the higher sample frequency of inevitable requirement, while generating the monitoring data of magnanimity, this
The real-time Transmission of a little data, synchronous storage are relatively difficult with post-processing.
Compressive sensing theory plays the role of enlightenment to the solution of the above problem, which points out, if original signal is dilute
It is thin or under certain transform domain it is compressible, can be with the rate far below Nyquist to being pressed while signal sampling
Contracting.Compressed sensing is mainly made of the acquisition of signal and reconstruction two large divisions, the measurement square as the most crucial content of compressed sensing
Battle array all plays an important role in this two parts：Calculation matrix performance is better, and the hits needed is fewer, and reconstruction error is also got over
It is small.Current calculation matrix is broadly divided into randomness matrix and certainty matrix two major classes.At compressive sensing theory initial stage, with height
This matrix is the randomness calculation matrix of representative, measures needed for it due to number is few, reconstruction accuracy is high and receives favor, but its structure
The memory space of complexity, occupancy is big, and random argument is more, does not utilize hardware realization.On the contrary, certainty calculation matrix is simple in structure,
Hardware construction difficulty is greatly reduced, Project Realization is conducive to, therefore, domestic and international many scholars transfer to have studied certainty square
Battle array, such as Toeplitz matrixes, cycle calculation matrix, but because the reconstruction error of certainty calculation matrix is larger, fail to obtain wide
General application.
For the problem that certainty calculation matrix reconstruction error in above-mentioned compressed sensing is larger, not yet propose at present effective
Solution.
Invention content
For the defect that certainty calculation matrix reconstruction error in the prior art is larger, propose a kind of based on certainty measurement
The vibration signal data of matrix compresses acquisition method, including：The original vibration signal for extracting acquisition, in dct basis
(DCT) sparsity analysis is carried out to original vibration signal under orthogonal basis, obtains degree of rarefication K；
According to degree of rarefication K, the first Symmetric Orthogonal certainty calculation matrix is established based on certainty calculation matrix；
The first Symmetric Orthogonal certainty calculation matrix of acquisition is iterated using threshold value iterative shrinkage algorithm, obtains
Two Symmetric Orthogonal certainty calculation matrix；
It is optimized using singular value decomposition algorithm pair the second Symmetric Orthogonal certainty calculation matrix, it is orthogonal right to obtain third
Claim certainty calculation matrix.
Further, in above-mentioned technical proposal, further include：
Compression measurement is carried out to original vibration signal according to third Symmetric Orthogonal certainty matrix and obtains measured value；
According to measured value, third Symmetric Orthogonal certainty calculation matrix and DCT orthogonal basis, sparse coefficient is reconstructed；
The vibration signal reconstructed according to sparse coefficient.
Further, in above-mentioned technical proposal, the first Symmetric Orthogonal certainty is established based on certainty calculation matrix and is measured
The step of matrix includes：
A sequence is selected inside from Bernoulli sequenceAfter sequence plus the element of N-1 flashback forms
First ray, the First ray form are (σ1,σ2,...,σN)=(γ, ε1,...,εN/2-1,β,εN/2-1,...,ε1), wherein N
For sequenceIn element number；
First ray is obtained into the second sequence by inverse Fourier transform, wherein the second sequence is that the first Symmetric Orthogonal is true
The first trip element of observational measurement matrix, the first trip element obtain the first Symmetric Orthogonal certainty calculation matrix by cyclic shift
Remaining other rows；
Randomly choose the M rows of the first Symmetric Orthogonal certainty calculation matrix, and multiplying factorStandardization obtains M × N the
One Symmetric Orthogonal certainty calculation matrix Φ, i.e.,Wherein, the relationship of M and degree of rarefication K need to meet formulaWherein c ≈ 0.28.
Further, the first certainty calculation matrix of acquisition is iterated using threshold value iterative shrinkage algorithm
Include to the step of the second Symmetric Orthogonal certainty calculation matrix：
To the first Symmetric Orthogonal certainty calculation matrix Φ arrange unitization, acquisition initial matrix Φ0；
According to degree of rarefication K,Determine Φ0Line number, i.e. measured value M；
Perception matrix D is sought according to DCT orthogonal basisq=ψDCTΦq, the perception matrix is carried out to arrange unitization obtainIts
In, ΦqFor Iterative Matrix, ψDCTFor DCT orthogonal basis, q is iterations；
According to the unitization perception matrix of rowUsing formulaObtain gram matrix, wherein T is to ask
The matrix turns order matrix；
The gram matrix G is updated according to threshold value t and scale descending factors γq, obtain updated gram matrix
Updated gram matrix is reduced using singular value decomposition algorithmOrder to M；
It enablesReverse goes out to perceive matrix Dq(Dq∈RM×N) to get to the perception after threshold value iterative shrinkage
Matrix Dq；
WithFor target update Φq+1, i.e., Norm minimum when
For the second Symmetric Orthogonal certainty calculation matrix Φ '.
Further, using singular value decomposition algorithm to obtained the second Symmetric Orthogonal certainty calculation matrix Φ '
The step of optimizing, obtaining third Symmetric Orthogonal certainty calculation matrix include：
Second Symmetric Orthogonal certainty calculation matrix is passed through into formula Φ '=U Λ PTDiagonalization Decomposition is done,
Wherein, U ∈ RM×MWith P ∈ RN×NIt is orthogonal matrix, Λ ∈ RM×NIt is diagonal matrix, the element on the diagonal line of Λ is to survey
The singular value of moment matrix Φ '.Now Λ is limited：Wherein, as only retained the factor big M before absolute value on the positive diagonal lines of Λ,
It is remaining to be all set as 0, thenWherein Δ=diag (σ1,σ2,…,σM), third Symmetric Orthogonal certainty is obtained after optimization
Calculation matrix Φ "=U Λ ' PT。
Further, it according to measured value, third Symmetric Orthogonal certainty calculation matrix Φ " and DCT orthogonal basis, reconstructs dilute
OMP algorithms are used during sparse coefficient.
According to an aspect of the present invention, a kind of storage medium is additionally provided, which includes the program of storage,
In, which executes above-mentioned data compression acquisition method.
According to another aspect of the present invention, a kind of processor is additionally provided, the processor is for running program, wherein should
Program executes above-mentioned data compression acquisition method when running.
The advantages of technical solution using the present invention, is：Extraction vibration signal carries out sparsity and analyzes to obtain degree of rarefication K,
Certainty calculation matrix is orthogonalized and symmetrization according to degree of rarefication K, independent argument is reduced, constructs and be easy to hardware realization
The first Symmetric Orthogonal certainty calculation matrix.In order to improve the reconstruction accuracy of the first Symmetric Orthogonal certainty calculation matrix, from
Incoherence is set out, and threshold value iterative shrinkage algorithm and singular value decomposition algorithm are combined, and first, passes through threshold value iterative shrinkage
Algorithm optimization the first Symmetric Orthogonal certainty calculation matrix obtains the second Symmetric Orthogonal certainty calculation matrix, to reduce by first just
The coherence between symmetrical certainty calculation matrix and sparse basis is handed over, singular value decomposition algorithm is secondly used, further increases the
Two Symmetric Orthogonal certainty calculation matrix itself column vector independence, finally obtain the third Symmetric Orthogonal suitable for vibration signal
Certainty calculation matrix certainty calculation matrix.The vibration signal compression gathering algorithm speed of service proposed in this paper is fast, is improving
Computation complexity is greatly reduced while calculation matrix compression performance.
Description of the drawings
Fig. 1 is that a kind of vibration signal data based on certainty calculation matrix of the present invention compresses acquisition method realization determination
Property calculation matrix optimization the step of；
Fig. 2 is that a kind of vibration signal data based on certainty calculation matrix of the present invention compresses acquisition method realization vibration
The step of signal reconstruction.
Specific implementation mode
Technical solution of the present invention is described in detail below in conjunction with attached drawing, following embodiment is only used for clearer
Ground illustrates technical scheme of the present invention, therefore is intended only as example, and not intended to limit the protection scope of the present invention.
Fig. 1 is that a kind of vibration signal data based on certainty calculation matrix of the present invention compresses acquisition method realization determination
Property calculation matrix optimization the step of.
In conjunction with Fig. 1, a kind of vibration signal data compression acquisition method realization certainty survey based on certainty calculation matrix
The step of moment matrix optimizes, includes the following steps：
Step 102：The original vibration signal for extracting acquisition, shakes under dct basis (DCT) orthogonal basis to original
Dynamic signal carries out sparsity analysis, obtains degree of rarefication K；
Step 104：According to degree of rarefication K, the first Symmetric Orthogonal certainty calculation matrix is established based on certainty calculation matrix；
Step 106：It is changed to the first Symmetric Orthogonal certainty calculation matrix of acquisition using threshold value iterative shrinkage algorithm
In generation, obtains the second Symmetric Orthogonal certainty calculation matrix；
Step 108：It is optimized using singular value decomposition algorithm pair the second Symmetric Orthogonal certainty calculation matrix, obtains
Three Symmetric Orthogonal certainty calculation matrix.
Fig. 2 is a kind of vibration signal data compression acquisition based on third Symmetric Orthogonal certainty calculation matrix of the present invention
Method realizes the step of vibration signal reconstruction.
Further, in one embodiment, further include：
Step 202:Compression is carried out according to third Symmetric Orthogonal certainty matrix to original vibration signal to be measured
Value M；
Step 204:According to measured value M, third Symmetric Orthogonal certainty calculation matrix and DCT orthogonal basis ψDCT, reconstruct dilute
Sparse coefficient；
Step 206:The vibration signal reconstructed according to sparse coefficient.
Further, the first Symmetric Orthogonal certainty is established based on certainty calculation matrix according to an embodiment of the invention to survey
The step of moment matrix includes：
A sequence is selected inside from Bernoulli sequenceAfter sequence plus the element of N-1 flashback forms
First ray, the First ray form are (σ1,σ2,...,σN)=(γ, ε1,...,εN/2-1,β,εN/2-1,...,ε1), wherein N
For sequenceIn element number；
First ray is obtained into the second sequence by inverse Fourier transform, wherein the second sequence is that the first Symmetric Orthogonal is true
The first trip element of observational measurement matrix, the first trip element obtain the first Symmetric Orthogonal certainty calculation matrix by cyclic shift
Remaining other rows；
Randomly choose the M rows of the first Symmetric Orthogonal certainty calculation matrix, and multiplying factorStandardization obtains M × N the
One Symmetric Orthogonal certainty calculation matrix Φ, i.e.,Wherein, the relationship of M and degree of rarefication K need to meet formulaWherein c ≈ 0.28.
Further, first certainty of acquisition is surveyed using threshold value iterative shrinkage algorithm according to embodiments of the present invention
Moment matrix is iterated the step of obtaining the second Symmetric Orthogonal certainty calculation matrix and includes：
To the first Symmetric Orthogonal certainty calculation matrix Φ arrange unitization, acquisition initial matrix Φ0；
According to degree of rarefication K,Determine Φ0Line number, i.e. measured value M；
Perception matrix D is sought according to DCT orthogonal basisq=ψDCTΦq, the perception matrix is carried out to arrange unitization obtainIts
In, ΦqFor Iterative Matrix, ψDCTFor DCT orthogonal basis, q is iterations；
According to the unitization perception matrix of rowUsing formulaObtain gram matrix, wherein T is to ask
The matrix turns order matrix；
The gram matrix G is updated according to threshold value t and scale descending factors γq, obtain updated gram matrix
Updated gram matrix is reduced using singular value decomposition algorithmOrder to M；
It enablesReverse goes out to perceive matrix Dq(Dq∈RM×N) to get to the perception after threshold value iterative shrinkage
Matrix Dq；
WithFor target update Φq+1, i.e., Norm minimum when
For the second Symmetric Orthogonal certainty calculation matrix Φ '.
It is understood that threshold value iterative shrinkage algorithm optimization the first certainty calculation matrix Φ is subject to incoherence
Then：It is firstly introduced into mutual coherence factor μ, indicates the column vector and DCT orthogonal basis ψ of the first Symmetric Orthogonal certainty calculation matrix ΦDCT
The inner product maximum value of column vector, the numerical value is smaller, then irrelevant property is better.Definition perception matrix Dq=ψDCTΦq, to perceiving square
Battle array carries out row unitization processing, obtains a new matrixThe mutual coherence factor μ (D) for then perceiving matrix column vector is：
Wherein diWithIt is D respectivelyqWithColumn vector, enable gram matrix
It is then based on the matrix, mutual coherence factor μ may be defined as at this time：
In formulaIt is the factor in gram matrix, is the first Symmetric Orthogonal certainty calculation matrix Φ's
Column vector and DCT orthogonal basis ψDCTThe inner product of different lines indicates the maximum value of off diagonal element in gram matrix.On however,
The μ for stating two kinds of equivalent definitions can only portray local correlations, therefore use the average cross correlation coefficient μ based on threshold value tt(D), right
The mould of element takes average degree on off-diagonal of the gram matrix not less than threshold value t of perception matrix, i.e.,：
The target optimized at this time is to reduce the first Symmetric Orthogonal certainty calculation matrix Φ and DCT orthogonal basis ψDCTT- it is flat
Equal cross-correlation coefficient μt(D), to obtain the second Symmetric Orthogonal certainty calculation matrix Φ '
Compared to the first Symmetric Orthogonal certainty calculation matrix Φ, it is true to obtain the second Symmetric Orthogonal through above-mentioned threshold value iterative algorithm
Observational measurement matrix Φ ', reconstruction property has a degree of raising, but is also not up to best perceived effect, i.e., at this time
To matrix not be it is optimal, therefore, will continue in next step using singular value decomposition algorithm optimize the second Symmetric Orthogonal determination
Property calculation matrix Φ '.
Further, use singular value decomposition algorithm described second orthogonal to what is obtained according to the abovementioned embodiments of the present invention
The step of symmetrical certainty calculation matrix Φ ' is optimized, obtained third Symmetric Orthogonal certainty calculation matrix Φ " include：
Second Symmetric Orthogonal certainty calculation matrix Φ ' is passed through into formula Φ '=U Λ PTDiagonalization Decomposition is done,
Wherein, U ∈ RM×MWith P ∈ RN×NIt is orthogonal matrix, Λ ∈ RM×NIt is diagonal matrix, the element on the diagonal line of Λ is
The singular value of two Symmetric Orthogonal certainty calculation matrix Φ '.Now Λ is limited：Wherein, as only retained on the positive diagonal lines of Λ absolutely
The factor big to M before value, remaining is all set as 0, then
Wherein Δ=diag (σ1,σ2,…,σM), third Symmetric Orthogonal certainty calculation matrix Φ "=U is obtained after optimization
Λ'PT。
It is understood that the second Symmetric Orthogonal certainty calculation matrix Φ ' can increase through singular value decomposition algorithm
The minimum singular value of two Symmetric Orthogonal certainty calculation matrix Φ ', the minimum singular value is bigger, and third Symmetric Orthogonal certainty is surveyed
The column vector independence of moment matrix is stronger, property when third Symmetric Orthogonal certainty calculation matrix Φ " is perceived for Signal Compression
It can be also stronger.
Further, according to measured value M, third Symmetric Orthogonal certainty calculation matrix Φ " and DCT orthogonal basis ψDCT, q weights
OMP algorithms are used during building out sparse coefficient.
By experimental verification, the first Symmetric Orthogonal certainty calculation matrix measurement by optimization compares, singular value point
First Symmetric Orthogonal certainty calculation matrix reconstruction error of resolving Algorithm optimization will produce the effect of reduction, and threshold value iterative algorithm is excellent
Change the first Symmetric Orthogonal certainty calculation matrix to need using the time as cost while improving reconstruction accuracy, the present invention is by two kinds
Method combines obtained third Symmetric Orthogonal certainty calculation matrix Φ ", can reduce reconstruction error minimum, improve matching degree,
Therefore reconstruction performance is effectively improved.
In the specification of the present invention, numerous specific details are set forth.It is to be appreciated, however, that the embodiment of the present invention can be with
It puts into practice without these specific details.In some instances, well known method, structure and skill is not been shown in detail
Art, so as not to obscure the understanding of this description.
It should be understood by those skilled in the art that, embodiments herein can be provided as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, the application can be used in one or more wherein include computer usable program code computer
The computer program production implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The application is with reference to method, the flow of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions every first-class in flowchart and/or the block diagram
The combination of flow and/or box in journey and/or box and flowchart and/or the block diagram.These computer programs can be provided
Instruct the processor of all-purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine so that the instruction executed by computer or the processor of other programmable data processing devices is generated for real
The device for the function of being specified in present one flow of flow chart or one box of multiple flows and/or block diagram or multiple boxes.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that instruction generation stored in the computer readable memory includes referring to
Enable the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one box of block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device so that count
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, in computer or
The instruction executed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, computing device includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/
Or the forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable Jie
The example of matter.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology realizes information storage.Information can be computer-readable instruction, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storages, magnetic tape cassette, tape magnetic disk storage or other magnetic storage apparatus
Or any other non-transmission medium, it can be used for storage and can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
Including so that process, method, commodity or equipment including a series of elements include not only those elements, but also wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including element
There is also other identical elements in process, method, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application
Form.It is deposited moreover, the application can be used to can be used in the computer that one or more wherein includes computer usable program code
The shape for the computer program product implemented on storage media (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
It these are only embodiments herein, be not intended to limit this application.To those skilled in the art,
The application can have various modifications and variations.It is all within spirit herein and principle made by any modification, equivalent replacement,
Improve etc., it should be included within the scope of claims hereof.
Claims (8)
1. a kind of vibration signal data based on certainty calculation matrix compresses acquisition method, which is characterized in that including：
The original vibration signal for extracting acquisition carries out the original vibration signal under dct basis DCT orthogonal basis
Sparsity is analyzed, and degree of rarefication K is obtained；
According to the degree of rarefication K, the first Symmetric Orthogonal certainty calculation matrix is established based on certainty calculation matrix；
The first Symmetric Orthogonal certainty calculation matrix of acquisition is iterated using threshold value iterative shrinkage algorithm, obtains
Two Symmetric Orthogonal certainty calculation matrix；
The second Symmetric Orthogonal certainty calculation matrix is optimized using singular value decomposition algorithm, it is orthogonal right to obtain third
Claim certainty calculation matrix.
2. data compression acquisition method according to claim 1, which is characterized in that further include：
Compression measurement is carried out to original vibration signal according to the third Symmetric Orthogonal certainty matrix and obtains measured value；
According to the measured value, the third Symmetric Orthogonal certainty calculation matrix and the DCT orthogonal basis, sparse system is reconstructed
Number；
The vibration signal reconstructed according to the sparse coefficient.
3. data compression acquisition method according to claim 1, which is characterized in that described to be built based on certainty calculation matrix
The step of vertical first Symmetric Orthogonal certainty calculation matrix includes：
A sequence is selected inside from Bernoulli sequenceAfter sequence plus the element of N-1 flashback forms first
Sequence, the First ray form are (σ1,σ2,...,σN)=(γ, ε1,...,εN/2-1,β,εN/2-1,...,ε1)
Wherein, N is sequenceIn element number；
First ray is obtained into the second sequence by inverse Fourier transform, second sequence is that first Symmetric Orthogonal determines
Property calculation matrix first trip element, the first trip element obtains the first Symmetric Orthogonal certainty by cyclic shift and measures square
Remaining other rows of battle array；
Randomly choose the M rows of the first Symmetric Orthogonal certainty calculation matrix, and multiplying factorStandardization is obtaining M × N first just
Symmetrical certainty calculation matrix Φ is handed over, i.e.,The relationship of wherein M and degree of rarefication K need to meet formulaWherein c ≈ 0.28.
4. data compression acquisition method according to claim 1, which is characterized in that described to use threshold value iterative shrinkage algorithm
The step of obtaining the second Symmetric Orthogonal certainty calculation matrix is iterated to the first certainty calculation matrix of acquisition to wrap
It includes：
To the first Symmetric Orthogonal certainty calculation matrix Φ arrange unitization, acquisition initial matrix Φ0；
According to the degree of rarefication K,Determine Φ0Line number, i.e. measured value M；
Perception matrix D is sought according to the DCT orthogonal basisq=ψDCTΦq, the perception matrix is carried out to arrange unitization obtainIts
In, ΦqFor Iterative Matrix, ψDCTFor DCT orthogonal basis, q is iterations；
According to the unitization perception matrix of the rowUsing formulaObtain gram matrix, wherein T is to ask
The matrix turns order matrix；
The gram matrix G is updated according to threshold value t and scale descending factors γq, obtain updated gram matrix
The updated gram matrix is reduced using singular value decomposition algorithmOrder to M；
It enablesReverse goes out to perceive matrix Dq(Dq∈RM×N) to get to the perception matrix after threshold value iterative shrinkage
Dq；
WithFor target update Φq+1, i.e., Norm minimum whenFor institute
State the second Symmetric Orthogonal certainty calculation matrix Φ '.
5. data compression acquisition method according to claim 1, which is characterized in that using singular value decomposition algorithm to obtaining
The second Symmetric Orthogonal certainty calculation matrix Φ ' optimize, obtain third Symmetric Orthogonal certainty calculation matrix
Step includes：
Second Symmetric Orthogonal certainty calculation matrix is passed through into formula Φ '=U Λ PTDiagonalization Decomposition is done,
Wherein, U ∈ RM×MWith P ∈ RN×NIt is orthogonal matrix, Λ ∈ RM×NIt is diagonal matrix, the element on the diagonal line of Λ is to measure square
The singular value of battle array Φ '.Now Λ is limited：Only retain the factor big M before absolute value on the positive diagonal lines of Λ, remaining is all set as 0, i.e.,
Wherein Δ=diag (σ1,σ2,…,σM), third Symmetric Orthogonal certainty calculation matrix the Φ "=U is obtained after optimization
Λ'PT。
6. data compression acquisition method according to claim 1, which is characterized in that according to the measured value, the third
Symmetric Orthogonal certainty calculation matrix Φ " and the DCT orthogonal basis use OMP algorithms during reconstructing sparse coefficient.
7. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein described program right of execution
Profit requires the data compression acquisition method described in any one of 1.
8. a kind of processor, which is characterized in that the processor is for running program, wherein right of execution when described program is run
Profit requires the data compression acquisition method described in any one of 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810336121.1A CN108599773B (en) | 2018-04-16 | 2018-04-16 | Vibration signal data compression acquisition method based on deterministic measurement matrix |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810336121.1A CN108599773B (en) | 2018-04-16 | 2018-04-16 | Vibration signal data compression acquisition method based on deterministic measurement matrix |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108599773A true CN108599773A (en) | 2018-09-28 |
| CN108599773B CN108599773B (en) | 2020-04-10 |
Family
ID=63622421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810336121.1A Active CN108599773B (en) | 2018-04-16 | 2018-04-16 | Vibration signal data compression acquisition method based on deterministic measurement matrix |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108599773B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111447229A (en) * | 2020-03-27 | 2020-07-24 | 青岛海信网络科技股份有限公司 | Large-scale data acquisition method and device based on compressed sensing theory |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103700061A (en) * | 2013-12-30 | 2014-04-02 | 东北大学 | Compressive-sensing-based digital image watermark embedding and extraction method |
| CN105181122A (en) * | 2015-08-06 | 2015-12-23 | 兰州理工大学 | Mechanical vibration signal data compression and acquisition method |
| CN106301383A (en) * | 2016-08-25 | 2017-01-04 | 东北大学 | A kind of signal processing method based on compressed sensing |
-
2018
- 2018-04-16 CN CN201810336121.1A patent/CN108599773B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103700061A (en) * | 2013-12-30 | 2014-04-02 | 东北大学 | Compressive-sensing-based digital image watermark embedding and extraction method |
| CN105181122A (en) * | 2015-08-06 | 2015-12-23 | 兰州理工大学 | Mechanical vibration signal data compression and acquisition method |
| CN106301383A (en) * | 2016-08-25 | 2017-01-04 | 东北大学 | A kind of signal processing method based on compressed sensing |
Non-Patent Citations (1)
| Title |
|---|
| 倪加明: "压缩感知中矩阵优化与快速重构方法研究", 《中国优秀硕士学位论文全文数据库（电子期刊）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111447229A (en) * | 2020-03-27 | 2020-07-24 | 青岛海信网络科技股份有限公司 | Large-scale data acquisition method and device based on compressed sensing theory |
| CN111447229B (en) * | 2020-03-27 | 2022-03-11 | 青岛海信网络科技股份有限公司 | Large-scale data acquisition method and device based on compressed sensing theory |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108599773B (en) | 2020-04-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Compressed sensing and reconstruction with bernoulli matrices | |
| CN105181122B (en) | Mechanical oscillation signal data compression acquisition method | |
| Salvini et al. | Fast gain calibration in radio astronomy using alternating direction implicit methods: Analysis and applications | |
| CN103595414B (en) | Sparse sampling and signal compressive sensing reconstruction method | |
| US8756045B2 (en) | Method, device, computer program and computer program product for determining a representation of a signal | |
| Izadi et al. | A compressed-sensing-based compressor for ECG | |
| CN103337087A (en) | Compressive sensing reconstruction method based on pseudo-inverse adaptive matching pursuit | |
| Almasri et al. | Toward compressed sensing of structural monitoring data using discrete cosine transform | |
| Esmaeili et al. | A new generalized shrinkage conjugate gradient method for sparse recovery | |
| Yao et al. | Research of incoherence rotated chaotic measurement matrix in compressed sensing | |
| CN111046737A (en) | Efficient intelligent sensing acquisition method for microseism signal detection | |
| CN107547088A (en) | Enhanced self-adapted segmentation orthogonal matching pursuit method based on compressed sensing | |
| Qian et al. | DTAE: Deep tensor autoencoder for 3-D seismic data interpolation | |
| Jánosi et al. | Time-series analysis of transient chaos | |
| CN108886588A (en) | It is used to form the NI Vision Builder for Automated Inspection of the one-dimensional digital representation of low information content scene | |
| Isenko et al. | Where is my training bottleneck? hidden trade-offs in deep learning preprocessing pipelines | |
| CN108599773A (en) | A kind of vibration signal data compression acquisition method based on certainty calculation matrix | |
| CN103391099A (en) | Random sampler suitable for one-dimensional slowly varying signals | |
| CN108846430B (en) | Image signal sparse representation method based on multi-atom dictionary | |
| Ozturk et al. | Application of compressive sensing to refractivity retrieval using networked weather radars | |
| Yue et al. | A improved CoSaMP algorithm based on correlation coefficient for compressed sensing image reconstruction | |
| Stankovic et al. | Reconstruction of randomly sampled sparse signals using an adaptive gradient algorithm | |
| De Luca et al. | Towards a GPU parallel software for environmental data fitting | |
| Kasem et al. | DRCS-SR: Deep robust compressed sensing for single image super-resolution | |
| Wang et al. | A joint matrix minimization approach for seismic wavefield recovery |
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 |