CN103269223B - A kind of analog signal compressive sampling method - Google Patents
A kind of analog signal compressive sampling method Download PDFInfo
- Publication number
- CN103269223B CN103269223B CN201310158504.1A CN201310158504A CN103269223B CN 103269223 B CN103269223 B CN 103269223B CN 201310158504 A CN201310158504 A CN 201310158504A CN 103269223 B CN103269223 B CN 103269223B
- Authority
- CN
- China
- Prior art keywords
- matrix
- analog signal
- signal
- integrator
- compression
- 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
Links
Landscapes
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention provides a kind of analog signal compressive sampling method, framework basis proposes compression calculation matrix, by to compression sampling value sequence and compression calculation matrix synchronous conversion, remove matrix-vector relational expression correlation on the impact of tested analog signal reconstruction property, finally with this relational expression, tested analog signal is reconstructed, obtains the sample sequence of tested analog signal.Like this, when integrator carries out integration to the signal after demodulation in sampling time section, do not carry out reset processing, solve the incomplete Sampling of signal that integrator causes discharge time, improve the performance of analog information Transpression sampling system.
Description
Technical field
The invention belongs to high speed preiodic type technical field of signal sampling, more specifically say, relate to a kind of analog signal compressive sampling method that can reduce system difficulty.
Background technology
Compression sampling technology is a kind of lack sampling method based on compressive sensing theory.This technology utilizes the tested analog signal of preiodic type after Fourier transform, a small amount of frequency content is only had to have remarkable amplitude, the amplitude of the frequency content of the overwhelming majority is this sparse characteristic of zero, high speed PRBS is adopted to carry out random demodulation at frequency domain to measured signal, demodulated output signal is compressed with integrator, finally with low speed ADC, the signal after compression is sampled, can accurate reconstruction primary signal and tested analog signal by optimization algorithm.
Existing compression sampling technology realizes principle as shown in Figure 1, and the expression formula of m sampled value y [m] is:
In formula, x (t) is tested analog signal, p
ct () is pseudo random sequence, T
sfor the sampling period.
Existing compressive sampling method, is needed after each sample to be resetted to integrator by auxiliary circuit, avoids the information coupling between adjacent double sampling with this, but, adopt auxiliary circuit complicated to integrator reset realizing circuit.And the time needed for each integrator resets is unknown, therefore, must adopt the longer resetting time of a unification when resetting to it.Integrator is in reseting procedure simultaneously, cannot collect, will cause information leakage to the energy of tested analog signal.This impact is presented as the incomplete sampling to tested analog signal in sampled value, will cause the distortion of reconstruction signal when being reconstructed tested analog signal with this sampled value.
Summary of the invention
The object of the invention is to overcome the deficiencies in the prior art, a kind of analog signal compressive sampling method is provided, to solve the incomplete sampling effect that integrator causes resetting time, improve the performance of compression sampling signal reconstruction.
For realizing above object, analog signal compressive sampling method of the present invention, is characterized in that, comprise the following steps:
(1), tested analog signal carries out random demodulation with the pseudo random sequence with signal nyquist frequency through frequency mixer, signal after demodulation exports all will carry the spectrum information of signal over the entire frequency band, integrator realizes the compression to signal after demodulation, last to sample to integral output signal far below the sample rate of signal nyquist frequency, obtain compression sampling value sequence;
In sampling time section, when integrator carries out integration to the signal after demodulation, do not carry out reset processing;
(2), according to the mathematics behavior model of compression sampling system, the compression calculation matrix after the synchronous conversion of structure;
(3), the compression sampling value sequence obtained synchronously is converted, according to the compression calculation matrix after the synchronous conversion of structure, obtain the matrix-vector relational expression removing correlation, finally with this relational expression, tested analog signal is reconstructed, obtain the sample sequence of tested analog signal, complete the compression sampling to tested analog signal.
The object of the present invention is achieved like this:
Analog signal compressive sampling method of the present invention, framework basis proposes compression calculation matrix, by to compression sampling value sequence and compression calculation matrix synchronous conversion, remove matrix-vector relational expression correlation on the impact of tested analog signal reconstruction property, finally with this relational expression, tested analog signal is reconstructed, obtains the sample sequence of tested analog signal.Like this, when integrator carries out integration to the signal after demodulation in sampling time section, do not carry out reset processing, solve the incomplete Sampling of signal that integrator causes discharge time, improve the performance of analog information Transpression sampling system.
Accompanying drawing explanation
Fig. 1 is conventional compression sampling principle block diagram;
Fig. 2 is a kind of embodiment theory diagram of analog signal compressive sampling method of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described, so that those skilled in the art understands the present invention better.Requiring particular attention is that, in the following description, when perhaps the detailed description of known function and design can desalinate main contents of the present invention, these are described in and will be left in the basket here.
Fig. 2 is a kind of embodiment theory diagram of analog signal compressive sampling method of the present invention.
As shown in Figure 2, in the present embodiment, analog signal compressive sampling method of the present invention comprises the following steps:
Step ST1: obtain compression sampling value sequence.
Tested analog signal signal x (t) and the pseudo random sequence P with signal nyquist frequency
ct () adopts frequency mixer to carry out random demodulation.In the present embodiment, in order to meet circuit realizability and compressive sensing theory to the requirement compressing calculation matrix, pseudo random sequence P
ct () adopts Rider horse contract pseudo random sequence composition.
Integrator realizes the compression to signal after demodulation, and integrator can adopt voltage integrator to realize.In the present invention, at sampling time section 0 ~ mT
sin, when integrator carries out integration to the signal after demodulation, do not carry out reset processing.This and prior art reset processing, at each sampling period T
scarry out integration difference.
Last to sample to integral output signal far below the sample rate of signal nyquist frequency, obtain.Compared with prior art, the present invention does not need to carry out reset processing to integrator.
Compression sampling value sequence y [m] expression formula is:
Wherein, m=1,2 ..., M, M are compression sampling value sequence length.
Step ST2: the compression calculation matrix Φ after the synchronous conversion of structure.
In the present invention, being functionally equivalent to each sampling period compression sampling value summation of integrator, therefore, the matrix form of integrator can be expressed as:
In formula, Matrix C is that M × N ties up matrix, and q=N/M, N are for treating reconstruction signal length, and in matrix, the element of position of non-1 is 0.When N can not be divided exactly by M, adjacent rows shares the information of a sampled value, and in Matrix C, temporally length ratio relation adopts fraction representation.Such as:
Work as M=3, during N=15, the expression formula of Matrix C is:
Work as M=3, during N=14, the expression formula of Matrix C is:
Compression calculation matrix Φ after the synchronous conversion of structure is:
C in formula
1c
mfor the row vector of integrator Matrix C, C ' is the integrator matrix after synchronous conversion.P is pseudo random sequence P
ct N × N that () is formed as diagonal element ties up diagonal matrix, if pseudo random sequence is [ε
1ε
2ε
n], then vector element ε
ivalue be 1 or-1, and value distribution probability meet:
the matrix notation of P is:
In formula, the value of off-diagonal element is 0.
Step ST3: synchronously convert compression sampling value sequence, carries out tested analog signal reconstruct after obtaining the matrix-vector relational expression of removal correlation.
In analog signal compressive sampling method of the present invention, owing to not resetting to integrator, therefore, current compression sampled value contains the information of each sampling period compression sampling value, there is between sampled value very strong correlation, in order to remove the impact of this correlation on compression sampling systematic function, current sample values is adopted to deduct the synchronous conversion of adjacent sample values above.Sampled value sequences y after synchronous conversion and the compression calculation matrix Φ after synchronous conversion have the matrix-vector relational expression removing correlation as follows:
y=Φx=Φ·Ψα,(9)
Wherein, y=[y [1] y [2]-y [1] ... y [M]-y [M-1]], in formula, Ψ is that frequency-domain sparse represents base, be made up of discrete Fourier transform (DFT) vector, Ψ is that N × N ties up matrix, α is the sample sequence x waiting to reconstruct tested analog signal represents base Ψ conversion coefficient at frequency-domain sparse, and the sequence length of α and x is N.
By compressed sensing signal reconstruction algorithm, obtain conversion coefficient α, obtain the sample sequence x of tested analog signal x (t) finally by inverse fourier transform.
Tested analog signal is reconstructed and belongs to prior art, do not repeat them here.
The compression sampling value sequence that the present invention proposes with compress calculation matrix and synchronously convert, can correlation be reduced.
Compression calculation matrix coefficient correlation μ (Φ, Ψ) is defined as:
φ in formula
iwith ψ
ji-th row vector of difference matrix Φ and Ψ and a jth column vector.
Convert the probability that front coefficient correlation is less than constant u (u>0) satisfy condition assuming that compression sampling value sequence be synchronous with compression calculation matrix (Φ=CP):
Compression sampling value sequence is synchronous with compression calculation matrix (Φ=C ' P) to be converted the probability that afterwards coefficient correlation is less than constant u and satisfies condition:
P
1with p
2for being greater than the normal number of 0, then p
2>p
1, that is: after conversion coefficient correlation to be less than the probability of constant u larger.
Prove:
Coefficient correlation μ (Φ, Ψ) before synchronous conversion, due to Φ=CP, so coefficient correlation can be rewritten as μ (CP, Ψ)=μ (C, P Ψ), due to
Here
ε
kfor a kth diagonal element of matrix P, c
ifor i-th row vector of Matrix C,
for the conjugate transpose of row k i-th column element of Matrix C, ψ
kjfor the row k jth column element of matrix Ψ.Following relation is had by Hough fourth inequality:
For all constant u>0,1≤i, j≤N, with
associating circle of probability is:
So:
Coefficient correlation μ (Φ, Ψ) after proving by the same methods synchronously converts=μ (C'P, Ψ) meets following condition:
In formula
obviously, can obtain according to the definition of Matrix C and C '
Assuming that
With
Due to
so p
2>p
1, that is: after conversion, coefficient correlation is less than the probability of constant u more greatly, can reduce correlation.
Although be described the illustrative embodiment of the present invention above; so that those skilled in the art understand the present invention; but should be clear; the invention is not restricted to the scope of embodiment; to those skilled in the art; as long as various change to limit and in the spirit and scope of the present invention determined, these changes are apparent, and all innovation and creation utilizing the present invention to conceive are all at the row of protection in appended claim.
Claims (3)
1. an analog signal compressive sampling method, is characterized in that, comprises the following steps:
(1), tested analog signal carries out random demodulation with the pseudo random sequence with signal nyquist frequency through frequency mixer, signal after demodulation exports all will carry the spectrum information of signal over the entire frequency band, integrator realizes the compression to signal after demodulation, last to sample to integral output signal far below the sample rate of signal nyquist frequency, obtain compression sampling value sequence;
In sampling time section, when integrator carries out integration to the signal after demodulation, do not carry out reset processing;
(2) the compression calculation matrix, after the synchronous conversion of structure:
C in formula
1c
mfor the row vector of integrator Matrix C, C ' is the integrator matrix after synchronous conversion; P is pseudo random sequence P
ct N × N that () is formed as diagonal element ties up diagonal matrix, and pseudo random sequence is [ε
1ε
2ε
n], then vector element ε
ivalue be 1 or-1, and value distribution probability meet:
matrix P is:
In formula, the value of off-diagonal element is 0;
Integrator Matrix C is:
In formula, integrator Matrix C is that M × N ties up matrix, and q=N/M, N are for treating reconstruction signal length, and in matrix, the element of position of non-1 is 0; When N can not be divided exactly by M, adjacent rows shares the information of a sampled value, and in integrator Matrix C, temporally length ratio relation adopts fraction representation;
(3), the compression sampling value sequence obtained synchronously is converted, according to the compression calculation matrix after the synchronous conversion of structure, obtain the matrix-vector relational expression removing correlation, finally with this relational expression, tested analog signal is reconstructed, obtain the sample sequence of tested analog signal, complete the compression sampling to tested analog signal.
2. analog signal compressive sampling method according to claim 1, is characterized in that, the pseudo random sequence described in step (1) is Rider horse contract pseudo random sequence.
3. analog signal compressive sampling method according to claim 1, is characterized in that, the matrix-vector relational expression of the removal correlation described in step (3):
y=Φx=Φ·Ψα,
Wherein, sampled value sequence y=[y [1] y [2]-y [1] after synchronous conversion ... y [M]-y [M-1]], in formula, Ψ is that frequency-domain sparse represents base, be made up of discrete Fourier transform (DFT) vector, Ψ is that N × N ties up matrix, α is the sample sequence x waiting to reconstruct tested analog signal represents base Ψ conversion coefficient at frequency-domain sparse, and the sequence length of α and x is N.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310158504.1A CN103269223B (en) | 2013-05-02 | 2013-05-02 | A kind of analog signal compressive sampling method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310158504.1A CN103269223B (en) | 2013-05-02 | 2013-05-02 | A kind of analog signal compressive sampling method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103269223A CN103269223A (en) | 2013-08-28 |
CN103269223B true CN103269223B (en) | 2016-01-20 |
Family
ID=49012835
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310158504.1A Expired - Fee Related CN103269223B (en) | 2013-05-02 | 2013-05-02 | A kind of analog signal compressive sampling method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103269223B (en) |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104467858A (en) * | 2013-09-25 | 2015-03-25 | 中国科学院深圳先进技术研究院 | Time domain integration sampling method and sampling circuit |
CN103986559B (en) * | 2014-05-27 | 2017-02-08 | 东南大学 | Five-order circulation cumulant estimation algorithm for compressed sampling signals |
CN104065383A (en) * | 2014-06-23 | 2014-09-24 | 中国工程物理研究院电子工程研究所 | Analog information conversion method based on sampling control |
CN104515981A (en) * | 2014-12-08 | 2015-04-15 | 广西大学 | Radar signal processing method and device based on compressive sensing |
CN106105053A (en) * | 2015-02-28 | 2016-11-09 | 华为技术有限公司 | A kind of compressive sampling method and device |
CN104682964B (en) * | 2015-03-15 | 2017-10-24 | 西安电子科技大学 | A kind of half determines the building method of compressed sensing calculation matrix |
CN105471525B (en) * | 2015-11-12 | 2019-03-08 | 中国电子科技集团公司第四十一研究所 | A kind of four-way compressed sensing digital receiver signal processing method of vector network analyzer |
CN106656201B (en) * | 2016-12-23 | 2020-06-30 | 燕山大学 | Compression method based on amplitude-frequency characteristics of sampled data |
CN106850473B (en) * | 2016-12-27 | 2019-09-24 | 电子科技大学 | A kind of broadband compression sampling system based on random demodulation |
CN107302362B (en) * | 2017-06-14 | 2020-04-24 | 南京工业大学 | Signal sparse representation method based on affine scale steepest descent algorithm |
WO2020082316A1 (en) * | 2018-10-26 | 2020-04-30 | 深圳大学 | System and method for processing single sampling radar signal |
CN113640654B (en) * | 2021-07-30 | 2024-02-20 | 深圳速跃芯仪科技有限公司 | High-speed state analysis method and system |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101951270A (en) * | 2010-08-24 | 2011-01-19 | 哈尔滨工业大学深圳研究生院 | Compressively sampling and receiving system and method for impulse ultra-wideband signals |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8836557B2 (en) * | 2010-10-13 | 2014-09-16 | Technion Research & Development Foundation Ltd. | Sub-Nyquist sampling of short pulses |
-
2013
- 2013-05-02 CN CN201310158504.1A patent/CN103269223B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101951270A (en) * | 2010-08-24 | 2011-01-19 | 哈尔滨工业大学深圳研究生院 | Compressively sampling and receiving system and method for impulse ultra-wideband signals |
Non-Patent Citations (1)
Title |
---|
稀疏模拟信号压缩采样与重构算法研究;赵贻玖;《中国博士学位论文全文数据库 信息科技辑》;20121215(第12期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN103269223A (en) | 2013-08-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103269223B (en) | A kind of analog signal compressive sampling method | |
WO2015172661A1 (en) | Compression sensing reconstruction method for monitoring microgrid harmonic wave | |
CN102104396B (en) | Pulse UWB (Ultra Wide Band) communication system based on CS (Compressed Sensing) theory | |
CN110365437B (en) | Fast power spectrum estimation method based on sub-Nyquist sampling | |
CN103036573A (en) | compressed sensing signal acquisition method based on filtering | |
Wang et al. | Compressive sensing reconstruction for vibration signals based on the improved fast iterative shrinkage-thresholding algorithm | |
CN104052494A (en) | Signal reconstruction method for frequency domain sparse signals | |
CN104767535A (en) | Low-complexity massive sparse signal reconstruction method | |
CN105578480A (en) | Undersampling frequency spectrum perception pre-decision method orienting broadband modulation converter | |
CN103558498B (en) | Based on the insulator pollution flashover leakage current signal sparse representation method of wavelet analysis | |
CN103684472A (en) | Reconfiguration method of adaptive signal of 1-Bit sparse level based on compression perception | |
CN103957009A (en) | Method for compensating for low-pass filter of compressed sampling system | |
CN102571034B (en) | Random circulant matrix-based analog compressed sensing sampling method and system | |
CN104734791B (en) | FRI (final random inspection) based sparse multiband signal frequency spectrum locating method | |
CN103823133A (en) | On-line power quality monitoring system based on compression sensing | |
CN103036574A (en) | Self-check sparseness self-adaption matching pursuit arithmetic based on compressive sensing | |
Sun et al. | Compressed sensing data reconstruction using adaptive generalized orthogonal matching pursuit algorithm | |
CN103036576A (en) | Two-value sparse signal reconstruction algorithm based on compressive sensing theory | |
CN103744828B (en) | Measurement value missing compression sensing reconstruction method | |
CN104300988A (en) | Compressed sensing based signal processing method and device | |
CN106027061A (en) | Lamb wave compression sensing method based on adaptive observation matrix | |
Zhang et al. | Modulated unit-norm tight frames for compressed sensing | |
CN103456148B (en) | The method and apparatus of signal reconstruction | |
CN105099462A (en) | Signal processing method based on compressive sensing | |
CN104901705A (en) | Compressed sampling method for multiband signal |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160120 Termination date: 20190502 |