CN111929724A - Early warning-oriented seismic signal compressed sensing acquisition transmission method - Google Patents

Early warning-oriented seismic signal compressed sensing acquisition transmission method Download PDF

Info

Publication number
CN111929724A
CN111929724A CN202010739680.4A CN202010739680A CN111929724A CN 111929724 A CN111929724 A CN 111929724A CN 202010739680 A CN202010739680 A CN 202010739680A CN 111929724 A CN111929724 A CN 111929724A
Authority
CN
China
Prior art keywords
signal
seismic
strong
earthquake
early warning
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
Application number
CN202010739680.4A
Other languages
Chinese (zh)
Other versions
CN111929724B (en
Inventor
夏界宁
黄�俊
陈志高
陈玉秀
余子昂
周立
罗松
杨厚丽
李丹
陈智慧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan Institute Of Seismologic Instrument Co ltd
Institute of Earthquake of China Earthquake Administration
Original Assignee
Wuhan Institute Of Seismologic Instrument Co ltd
Institute of Earthquake of China Earthquake Administration
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan Institute Of Seismologic Instrument Co ltd, Institute of Earthquake of China Earthquake Administration filed Critical Wuhan Institute Of Seismologic Instrument Co ltd
Priority to CN202010739680.4A priority Critical patent/CN111929724B/en
Publication of CN111929724A publication Critical patent/CN111929724A/en
Application granted granted Critical
Publication of CN111929724B publication Critical patent/CN111929724B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/28—Processing seismic data, e.g. analysis, for interpretation, for correction
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/10—Alarms for ensuring the safety of persons responsive to calamitous events, e.g. tornados or earthquakes
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Abstract

The invention provides an early warning-oriented seismic signal compressed sensing acquisition transmission method, which comprises the steps of seismic pickup, strong seismic signal framing processing, strong seismic signal compressed sampling, observed value output, signal reconstruction and output; the method evaluates the reconstructed signal at regular intervals, selects a corresponding observation matrix and a sparse basis at proper time, and improves the reliability of signal adoption and reconstruction; the method saves the resource consumption of the system at the data acquisition end and the communication end while compressing the redundant data, and improves the early warning efficiency.

Description

Early warning-oriented seismic signal compressed sensing acquisition transmission method
Technical Field
The invention relates to the technical field of seismic signal detection and early warning, in particular to an early warning-oriented seismic signal compressed sensing acquisition transmission method.
Background
One of natural disasters seriously threatening the safety of human life and property during earthquake often causes serious building damage and personal casualties, and causes fire, flood, pipeline leakage, diffusion of bacteria and radioactive substances, and secondary disasters such as landslide, tsunami, collapse, ground cracks and the like. After a large earthquake occurs, a large amount of energy is released in a short time, and destructive damage is caused within dozens of seconds or even seconds. With the development of global urbanization, cities gradually become centers of political, economic, cultural and social activities, and the cities are high in population density and large in buildings, so that sudden earthquakes pose serious threats to the cities. The occurrence or non-occurrence of earthquakes is not controlled by human beings, which is a common feature of earthquakes and many other natural disasters. Due to the extremely complex process of induction and occurrence of the earthquake, reliable short-term earthquake prediction cannot be obtained at present. All countries in the world have put the emphasis on earthquake disaster reduction measures in the aspects of improving the earthquake resistance of buildings, earthquake early warning, earthquake emergency rescue and the like.
After the earthquake occurs, if the final scale of the earthquake can be identified by utilizing P waves at the early stage of fault fracture, earthquake early warning information is sent to an early warning target area about to be attacked by the earthquake through electric signals before destructive S waves arrive. Therefore, the areas receiving the early warning information can take disaster reduction measures in advance. The earthquake early warning can provide early warning time of several seconds to tens of seconds after an earthquake occurs and before destructive earthquake waves arrive, and corresponding emergency treatment measures are taken, so that the earthquake early warning is an effective means for lightening earthquake loss, reducing secondary disasters of the earthquake and reducing casualties newly developed in recent years. At present, relatively mature earthquake early warning systems are established in Japan, America, Mexico, Turkey, China and the like.
The earthquake early warning system is mainly implemented through the steps of data acquisition, data transmission, data processing, analysis and judgment and the like, earthquake data observed in the data acquisition step are transmitted to the data processing step for processing, noise interference is weakened, the signal-to-noise ratio and the resolution ratio of earthquake signals are improved, and the distortion degree of the earthquake signals is reduced, so that accurate early warning results can be obtained. As the number of observation station networks of the earthquake early warning system is continuously increased, the acquired data is increased in a geometric progression, the earthquake early warning has higher requirement on the timeliness of the data, and a contradiction exists between the acquired data and the data transmission speed. The common data acquisition of the earthquake platform network is based on the classic Nyquist sampling theorem; in view of data processing flow, the existing earthquake observation method is to transmit real-time sampling data after complete acquisition and recording, redundancy exists in the acquired data, a large amount of system resources are consumed in the acquisition process, and the acquired data is compressed and transmitted, so that equivalent system resources are consumed, and great pressure is brought to data recording and real-time transmission. If the compression of the signal is performed while sampling, the acquisition, transmission and reconstruction of the signal can be realized by consuming lower system resources.
Disclosure of Invention
In view of the above, the invention provides an early warning-oriented seismic signal compressed sensing acquisition transmission method which compresses seismic signals while acquiring the seismic signals, reduces redundancy in acquired data, and reliably restores the seismic signals.
The technical scheme of the invention is realized as follows: the invention provides an early warning-oriented seismic signal compressed sensing acquisition transmission method, which comprises the following steps:
s1: the earthquake early warning system carries out framing processing on an original strong earthquake signal containing noise in the acquired signal, the calculation efficiency is considered, meanwhile, the signal recovery effect is improved, adjacent frames are not overlapped, and a strong earthquake signal x after framing processing is obtained;
s2: carrying out compression sampling on the strong seismic signal x after framing processing, and constructing a strong seismic signal x model based on compressed sensing: let Ψ be a sparse group; Ψ' is the transpose of the sparse basis Ψ; x is a noise-containing framing original strong seismic signal, and x is not sparse; a is a sparse representation vector of x at a sparse basis Ψ, a is sparse, and a ═ Ψ' x; Φ is the observation matrix unrelated to the sparse basis Ψ, then:
y Φ a Φ Ψ' x; y is a low-dimensional observation vector of x under the observation matrix phi;
s3: after the low-dimensional observation vector y is obtained after compression sampling is completed, the low-dimensional observation vector y is sent into a communication sending buffer of the earthquake early warning system and is stored into a receiving buffer at a receiving end; every K frames, the sending buffer of the earthquake early warning system also sends the original strong earthquake signal after frame division processing to the receiving buffer of the system;
s4: receiving the compressed and sampled strong earthquake signal and the original strong earthquake signal by the signal, and reconstructing the compressed and sampled strong earthquake signal; the reconstruction algorithm is as follows:
according to the observation matrixes phi and psi', which are the transposed matrixes of sparse bases psi, the framed original strong seismic signals x containing noise and the compressed observation vectors y; phi psi' is used as a recovery matrix, and an orthogonal matching pursuit algorithm is adopted to reconstruct a compressed M-dimensional observation vector y of linear sampling to obtain an approximation signal of a strong seismic signal x after framing processing
Figure BDA0002606341880000031
Within the allowable error range, solving the minimum L0Norm:
Figure BDA0002606341880000032
s.t | y-phi Ψ' | is less than or equal to; wherein | x | purple0L being x0A norm; is a set error constant;
s5: approximating the reconstructed strong seismic signal xNumber (C)
Figure BDA0002606341880000033
Outputting the strong vibration signal after the noise reduction processing of the original strong vibration signal x containing noise; repeating the steps S2-S5 again until the output of the strong shock signals of all the sub-frames is finished;
s6: every K frames, approximation signals of the reconstructed strong seismic signals
Figure BDA0002606341880000034
And evaluating, and selectively adjusting the sparse basis psi and the observation matrix phi when processing the next adjacent frame length according to the evaluation result.
Based on the above technical solution, preferably, the framing process is to set the number of data points in each frame, the number of data points in each frame is one of 200, 400, 600, 512, or 800, and the sampling rate per second is set to be 100sps or 200 sps.
Further preferably, the sparse basis Ψ is a DCT, DFT, or wavelet DWT basis.
Still further preferably, the observation matrix Φ is a random gaussian matrix, a random bernoulli matrix, a sparse random measurement matrix, a partial fourier matrix, or a partial hadamard matrix.
Still further preferably, the pair of reconstructed strong seismic signals
Figure BDA0002606341880000035
The evaluation method comprises the following steps of measuring the quality of a reconstructed signal by sampling a signal-to-noise ratio (SNR) and a mean square error (MES):
Figure BDA0002606341880000036
Figure BDA0002606341880000037
wherein N is the length of the sub-frame and represents the number of the strong seismic data in each sub-frame; measuring the quantity M, i.e. pressureThe observation number, x, of the sampled strong seismic signal is reducediA frame of strong earthquake signals which are sent by the earthquake early warning system and are subjected to framing processing;
Figure BDA0002606341880000041
the reconstructed strong seismic signal is obtained; i | · | purple wind2Is the norm of L2; the compression ratio is M/N; under the condition of the same compression ratio, sampling different sparse bases psi and an observation matrix phi to calculate a signal-to-noise ratio SNR and a mean square error MES; and adjusting the compression ratio M/N, continuously calculating the signal-to-noise ratio SNR and the mean square error MES, observing whether the signal-to-noise ratio SNR or the mean square error MES changes correspondingly along with the change of the compression ratio, and when the signal-to-noise ratio SNR is maximum and the mean square error MES is minimum, indicating that the signal reconstruction and recovery effect under the sparse basis psi and the observation matrix phi is best.
On the basis of the above technical solution, preferably, the method further includes a signal pickup step, and the signal pickup method includes: the earthquake early warning system carries out 0.1-33 Hz band-pass filtering on the acquired earthquake P wave signals before framing processing, and adopts an STA/LTA method to roughly pick up the earthquake P waves after band-pass filtering:
STA is a short time window, LTA is a long time window; when the P wave of the seismic signal arrives, the STA/LTA has sudden change, and when the ratio of the STA/LTA to the LTA is greater than a set threshold R, the STA/LTA is judged to be an effective seismic signal:
Figure BDA0002606341880000042
CFPfor pick-up characteristic functions of P-waves of seismic signals, CFP＝Y(k)2+[Y(k)-Y(k-1)]2(ii) a Y (k) and Y (k-1) are acceleration recorded values in the vertical direction at the moment of k and k-1 respectively; i is the current moment; k is a radical of1And k2Is a time between the current times, and k2＜k1< i; in the above formula, the number of data points in the short time window of the STA of the numerator portion of the formula is 50, and the number of data points in the long time window of the LTA of the numerator portion of the formula is 1000;
the arrival time of the seismic phase of the seismic P wave can only be roughly identified by applying the STA/LTA method, and in order to accurately pick up the arrival time of the seismic phase of the seismic P wave, a VAR-AIC method is adopted for accurately picking up:
AIC(n)＝n·log{var(x[1,n])}+(L-n-1)·log{var(x[n+1,L])}；
the range of n is all sampling points in the seismic P wave STA or LTA window; var represents the variance; and the minimum value of the AIC in the window is a time-reaching point, and then P-wave data roughly picked up by adopting an STA/LTA method is pushed forward by the corresponding point number of the minimum value of the AIC or corresponding sampling time according to the acquisition time to realize accurate signal pickup.
Compared with the prior art, the earthquake signal compressed sensing acquisition transmission method facing the early warning has the following beneficial effects:
(1) according to the invention, sparse representation of non-sparse signals is realized by setting sparse basis, observation matrix, compression sampling model and reconstruction algorithm, the redundancy of sampling data volume is reduced, the reliability and reproducibility of original signal reconstruction are ensured, the system overhead is saved, and the signal processing efficiency is improved;
(2) the signals are subjected to framing in advance, so that the data processing amount can be reduced, and sparse bases and observation matrixes adopted by subsequent framing signals can be adjusted in time in the subsequent evaluation process, and the precision of signal compression and reconstruction is improved in time;
(3) the signal-to-noise ratio and the mean square error of the original strong seismic signals and the reconstructed strong seismic signals are evaluated regularly, the most suitable sparse basis and the observation matrix can be selected in time for signal compression and reconstruction, and the reliability of the signals sampled in real time is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flow chart of a seismic signal compressed sensing acquisition transmission method for early warning according to the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
As shown in fig. 1, the invention provides a seismic signal compressed sensing acquisition transmission method facing early warning, which specifically comprises the following steps:
s1: the earthquake early warning system carries out framing processing on an original strong earthquake signal containing noise in the acquired signal, the calculation efficiency is considered, meanwhile, the signal recovery effect is improved, adjacent frames are not overlapped, and a strong earthquake signal x after framing processing is obtained; the framing processing is to set the number of data points in each frame, the number of data points in each frame is one of 200, 400, 600, 512 or 800, the sampling rate per second is set to be 100sps or 200sps, namely the sampling number per second, if the sampling rate is 100sps and the number of framing points is 400, the sampling time is 4 seconds, and the time interval between adjacent points is 0.01 second; adjacent frames are not overlapped, and can be continuously framed or alternatively framed;
s2: carrying out compression sampling on the strong seismic signal x after framing processing, and constructing a strong seismic signal x model based on compressed sensing: let Ψ be a sparse group; Ψ' is the transpose of the sparse basis Ψ; x is a noise-containing framing original strong seismic signal, and x is not sparse; a is a sparse representation vector of x at a sparse basis Ψ, a is sparse, and a ═ Ψ' x; Φ is the observation matrix unrelated to the sparse basis Ψ, then:
y Φ a Φ Ψ' x; y is a low-dimensional observation vector of x under the observation matrix phi;
where the sparse basis Ψ may be a DCT, DFT, or wavelet DWT basis:
DFT is discrete fourier transform, which relates the time domain to the frequency domain, and transforms the time-transformed signal into fourier transform in the frequency domain;
DCT is a discrete cosine transform, similar to discrete fourier transform, but only uses real numbers. The real signals are obtained in the frequency domain after discrete cosine transform, and compared with DFT, the calculation amount can be reduced by half; DCT has strong energy concentration function, and the energy is mainly concentrated on low-frequency components;
the wavelet DWT base overcomes the defect that the traditional Fourier transform can not express the local characteristics of signals, better solves the contradiction between time resolution and frequency resolution, and has different resolutions at different positions of a time-frequency plane;
the observation matrix needs to be stable enough, in order to ensure that the linear observation of the signal can keep the original structure of the signal, the observation matrix must meet a constraint isometry property condition (RIP), and the observation matrix phi of the invention is a random Gaussian matrix, a random Bernoulli matrix, a sparse random measurement matrix, a partial Fourier matrix or a partial Hadamard matrix which meets the constraint isometry property condition RIP:
elements in the random Gaussian matrix obey random independent normal distribution in a (0, 1) interval, so that the elements are uncorrelated with most orthogonal bases (sparse bases) and meet constraint equidistant conditions;
each element of the random Bernoulli matrix independently obeys symmetrical Bernoulli distribution, and relative to the random Gaussian measurement matrix, the elements of the random Bernoulli measurement matrix are +/-1, so that the random Bernoulli measurement matrix is easier to realize and store in practical application;
the construction method of the sparse random measurement matrix is as follows: firstly, generating an all-zero matrix phi with the size of M multiplied by N, wherein M is less than N, and then randomly selecting K positions to be set as 1 for each row in the matrix phi, wherein K is less than M; each column of the sparse random measurement matrix only has K nonzero elements, the structure is simple, and the sparse random measurement matrix is easy to construct and store in practical application;
part of Fourier matrix belongs to part of orthogonal matrix, satisfies constraint equidistant condition (RIP), and can be used as observation matrix; firstly, generating a Fourier matrix with the size of NxN, then randomly selecting M row vectors in the generated matrix to form an MxN matrix, and finally normalizing the columns; similar to the partial Fourier matrix, the partial Hadamard matrix also belongs to a partial orthogonal matrix, and has stronger non-correlation and partial orthogonality, so that compared with other deterministic measurement matrices, the measurement matrix has fewer measurement times required for accurate reconstruction and good reconstruction effect. However, the dimension N must be an integral multiple of 2, which limits the application range and the occasion of the matrix to a certain extent;
the optimal observation matrix phi is a random Gaussian matrix, a random Bernoulli matrix or a sparse random measurement matrix;
s3: after the low-dimensional observation vector y is obtained after compression sampling is completed, the low-dimensional observation vector y is sent into a sending buffer of the earthquake early warning system and is sent to a receiving buffer of the system; every K frames, the sending buffer of the earthquake early warning system also sends a frame of strong earthquake signals which are subjected to framing processing to a receiving buffer of the system;
s4: receiving the compressed and sampled strong earthquake signal and the original strong earthquake signal by the signal, and reconstructing the compressed and sampled strong earthquake signal; the reconstruction algorithm is as follows:
according to the observation matrixes phi and psi', which are the transposed matrixes of sparse bases psi, the framed original strong seismic signals x containing noise and the compressed observation vectors y; phi psi' is used as a recovery matrix, and an orthogonal matching pursuit algorithm is adopted to reconstruct a compressed M-dimensional observation vector y of linear sampling to obtain an approximation signal of a strong seismic signal x after framing processing
Figure BDA0002606341880000071
Within the allowable error range, solving the minimum L0Norm:
Figure BDA0002606341880000072
s.t | y-phi Ψ' | is less than or equal to; wherein | x | purple0L being x0A norm; is a set error constant;
s5: approximating the reconstructed strong seismic signal x to a signal
Figure BDA0002606341880000081
Outputting the strong vibration signal after the noise reduction processing of the original strong vibration signal x containing noise; repeating the steps S2-S5 again until the output of the strong shock signals of all the sub-frames is finished;
s6: every K frames, approximation signals of the reconstructed strong seismic signals
Figure BDA0002606341880000082
Evaluating, and selectively adjusting sparse basis psi and observation matrix phi when processing the next adjacent frame length according to the evaluation result;
for the reconstructed strong seismic signal
Figure BDA0002606341880000083
The specific evaluation method comprises the following steps: the reconstructed signal quality is measured by sampling the signal-to-noise ratio SNR and the mean square error MES:
Figure BDA0002606341880000084
Figure BDA0002606341880000085
wherein N is the length of the sub-frame and represents the number of the strong seismic data in each sub-frame; the number of measurements M, i.e. the number of observations, x, of the macroseism signal after compression samplingiA frame of strong earthquake signals which are sent by the earthquake early warning system and are subjected to framing processing;
Figure BDA0002606341880000086
the reconstructed strong seismic signal is obtained; i | · | purple wind2Is the norm of L2; the compression ratio is M/N; under the condition of the same compression ratio, sampling different sparse bases psi and an observation matrix phi to calculate a signal-to-noise ratio SNR and a mean square error MES; adjusting the compression ratio M/N, continuously calculating the signal-to-noise ratio SNR and the mean square error MES, observing whether the signal-to-noise ratio SNR or the mean square error MES changes correspondingly along with the change of the compression ratio, and expressing the signal reconstruction and the recovery effect under the sparse basis psi and the observation matrix phi when the signal-to-noise ratio SNR value is maximum and the mean square error MES is minimumThe best result is obtained.
In order to further reduce the signal processing amount of the earthquake early warning system and accurately identify the strong earthquake signal from the detection signal, the method of the invention also comprises a signal pickup step, and the signal pickup method comprises the following steps: before framing processing, the earthquake early warning system carries out 0.1-33 Hz band-pass filtering on the acquired earthquake P wave signals, and adopts an STA/LTA method to roughly pick up the earthquake P waves after band-pass filtering:
STA is a short time window, LTA is a long time window; when the P wave of the seismic signal arrives, the STA/LTA has sudden change, and when the ratio of the STA/LTA to the LTA is greater than a set threshold R, the STA/LTA is judged to be an effective seismic signal:
Figure BDA0002606341880000091
CFPfor pick-up characteristic functions of P-waves of seismic signals, CFP＝Y(k)2+[Y(k)-Y(k-1)]2(ii) a Y (k) and Y (k-1) are acceleration recorded values in the vertical direction at the moment of k and k-1 respectively; i is the current moment; k is a radical of1And k2Is a time between the current times, and k2＜k1< i; in the above formula, the number of data points in the short time window of the STA of the numerator portion of the formula is 50, and the number of data points in the long time window of the LTA of the numerator portion of the formula is 1000;
the arrival time of the seismic phase of the seismic P wave can only be roughly identified by applying the STA/LTA method, and in order to accurately pick up the arrival time of the seismic phase of the seismic P wave, a VAR-AIC method is adopted for accurately picking up:
AIC(n)＝n·log{var(x[1,n])}+(L-n-1)·log{var(x[n+1,L])}；
the range of n is all sampling points in the seismic P wave STA or LTA window; var represents the variance; and the minimum value of the AIC in the window is a time-reaching point, and then P-wave data roughly picked up by adopting an STA/LTA method is pushed forward by the corresponding point number of the minimum value of the AIC or corresponding sampling time according to the acquisition time to realize accurate signal pickup. Because of the sampling rate per second, the forward movement of the sampling points corresponds to the forward movement of the corresponding sampling time, so that the acquired data is more reliable in timeliness.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (6)

1. A seismic signal compressed sensing acquisition transmission method facing early warning is characterized in that: the method comprises the following steps:
s1: the earthquake early warning system carries out framing processing on an original strong earthquake signal containing noise in the acquired signal, the calculation efficiency is considered, meanwhile, the signal recovery effect is improved, adjacent frames are not overlapped, and a strong earthquake signal x after framing processing is obtained;
s2: carrying out compression sampling on the strong seismic signal x after framing processing, and constructing a strong seismic signal x model based on compressed sensing: let Ψ be a sparse group; Ψ' is the transpose of the sparse basis Ψ; x is a noise-containing framing original strong seismic signal, and x is not sparse; a is a sparse representation vector of x at a sparse basis Ψ, a is sparse, and a ═ Ψ' x; Φ is the observation matrix unrelated to the sparse basis Ψ, then:
y Φ a Φ Ψ' x; y is a low-dimensional observation vector of x under the observation matrix phi;
s3: after the low-dimensional observation vector y is obtained after compression sampling is completed, the low-dimensional observation vector y is sent into a communication sending buffer of the earthquake early warning system and is stored into a receiving buffer at a receiving end; every K frames, the sending buffer of the earthquake early warning system also sends the original strong earthquake signal after frame division processing to the receiving buffer of the system;
s4: receiving the compressed and sampled strong earthquake signal and the original strong earthquake signal by the signal, and reconstructing the compressed and sampled strong earthquake signal; the reconstruction algorithm is as follows:
according to the observation matrixes phi and psi', which are the transposed matrixes of sparse bases psi, the framed original strong seismic signals x containing noise and the compressed observation vectors y; phi psi' is used as a recovery matrix, and an orthogonal matching pursuit algorithm is adopted to reconstruct a compressed M-dimensional observation vector y of linear sampling to obtain an approximation signal of the strong seismic signal x after framing processingNumber (C)
Figure FDA0002606341870000011
Within the allowable error range, solving the minimum L0Norm:
Figure FDA0002606341870000012
s.t | y-phi Ψ' | is less than or equal to; wherein | x | purple0L being x0A norm; is a set error constant;
s5: approximating the reconstructed strong seismic signal x to a signal
Figure FDA0002606341870000013
Outputting the strong vibration signal after the noise reduction processing of the original strong vibration signal x containing noise; repeating the steps S2-S5 again until the output of the strong shock signals of all the sub-frames is finished;
s6: every K frames, approximation signals of the reconstructed strong seismic signals
Figure FDA0002606341870000014
And evaluating, and selectively adjusting the sparse basis psi and the observation matrix phi when processing the next adjacent frame length according to the evaluation result.
2. The compressed sensing acquisition and transmission method for the early warning seismic signals as claimed in claim 1, wherein the compressed sensing acquisition and transmission method comprises the following steps: the framing process is to set the number of data points in each frame to be one of 200, 400, 600, 512, or 800, and to set the sampling rate per second to be 100sps or 200 sps.
3. The compressed sensing acquisition and transmission method for the early warning seismic signals as claimed in claim 2, wherein the compressed sensing acquisition and transmission method comprises the following steps: the sparse basis Ψ is a DCT, DFT, or wavelet DWT basis.
4. The compressed sensing acquisition and transmission method for the early warning seismic signals as claimed in claim 3, wherein the compressed sensing acquisition and transmission method comprises the following steps: the observation matrix phi is a random Gaussian matrix, a random Bernoulli matrix, a sparse random measurement matrix, a partial Fourier matrix or a partial Hadamard matrix.
5. The compressed sensing acquisition and transmission method for the early warning seismic signals as claimed in claim 3, wherein the compressed sensing acquisition and transmission method comprises the following steps: the pair of reconstructed strong seismic signals
Figure FDA0002606341870000021
The evaluation method comprises the following steps of measuring the quality of a reconstructed signal by sampling a signal-to-noise ratio (SNR) and a mean square error (MES):
Figure FDA0002606341870000022
Figure FDA0002606341870000023
wherein N is the length of the sub-frame and represents the number of the strong seismic data in each sub-frame; the number of measurements M, i.e. the number of observations, x, of the macroseism signal after compression samplingiA frame of strong earthquake signals which are sent by the earthquake early warning system and are subjected to framing processing;
Figure FDA0002606341870000024
the reconstructed strong seismic signal is obtained; i | · | purple wind2Is the norm of L2; the compression ratio is M/N; under the condition of the same compression ratio, sampling different sparse bases psi and an observation matrix phi to calculate a signal-to-noise ratio SNR and a mean square error MES; and adjusting the compression ratio M/N, continuously calculating the signal-to-noise ratio SNR and the mean square error MES, observing whether the signal-to-noise ratio SNR or the mean square error MES changes correspondingly along with the change of the compression ratio, and when the signal-to-noise ratio SNR is maximum and the mean square error MES is minimum, indicating that the signal reconstruction and recovery effect under the sparse basis psi and the observation matrix phi is best.
6. The compressed sensing acquisition and transmission method for the early warning seismic signals as claimed in claim 1, wherein the compressed sensing acquisition and transmission method comprises the following steps: the method also comprises a signal pickup step, and the signal pickup method comprises the following steps: the earthquake early warning system carries out 0.1-33 Hz band-pass filtering on the acquired earthquake P wave signals before framing processing, and adopts an STA/LTA method to roughly pick up the earthquake P waves after band-pass filtering:
STA is a short time window, LTA is a long time window; when the P wave of the seismic signal arrives, the STA/LTA has sudden change, and when the ratio of the STA/LTA to the LTA is greater than a set threshold R, the STA/LTA is judged to be an effective seismic signal:
Figure FDA0002606341870000031
CFPfor pick-up characteristic functions of P-waves of seismic signals, CFP＝Y(k)2+[Y(k)-Y(k-1)]2(ii) a Y (k) and Y (k-1) are acceleration recorded values in the vertical direction at the moment of k and k-1 respectively; i is the current moment; k is a radical of1And k2Is a time between the current times, and k2＜k1< i; in the above formula, the number of data points in the short time window of the STA of the numerator portion of the formula is 50, and the number of data points in the long time window of the LTA of the numerator portion of the formula is 1000;
the arrival time of the seismic phase of the seismic P wave can only be roughly identified by applying the STA/LTA method, and in order to accurately pick up the arrival time of the seismic phase of the seismic P wave, a VAR-AIC method is adopted for accurately picking up:
AIC(n)＝n·log{var(x[1,n])}+(L-n-1)·log{var(x[n+1,L])}；
the range of n is all sampling points in the seismic P wave STA or LTA window; var represents the variance; and the minimum value of the AIC in the window is a time-reaching point, and then P-wave data roughly picked up by adopting an STA/LTA method is pushed forward by the corresponding point number of the minimum value of the AIC or corresponding sampling time according to the acquisition time to realize accurate signal pickup.
CN202010739680.4A 2020-07-28 2020-07-28 Earthquake signal compressed sensing acquisition transmission method facing early warning Active CN111929724B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010739680.4A CN111929724B (en) 2020-07-28 2020-07-28 Earthquake signal compressed sensing acquisition transmission method facing early warning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010739680.4A CN111929724B (en) 2020-07-28 2020-07-28 Earthquake signal compressed sensing acquisition transmission method facing early warning

Publications (2)

Publication Number Publication Date
CN111929724A true CN111929724A (en) 2020-11-13
CN111929724B CN111929724B (en) 2022-05-31

Family

ID=73314796

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010739680.4A Active CN111929724B (en) 2020-07-28 2020-07-28 Earthquake signal compressed sensing acquisition transmission method facing early warning

Country Status (1)

Country Link
CN (1) CN111929724B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112965101A (en) * 2021-04-25 2021-06-15 福建省地震局应急指挥与宣教中心 Earthquake early warning information processing method
CN113573273A (en) * 2021-08-02 2021-10-29 尧米(重庆)智能科技有限公司 Wireless intelligent acquisition synchronization method and system based on Bluetooth transmission
CN114647006A (en) * 2022-05-18 2022-06-21 东营市京川石油技术有限公司 Earthquake weak signal enhancement method based on multiple compressed sensing
CN115996144A (en) * 2022-12-08 2023-04-21 西南石油大学 Fracturing data transmission method and system based on compressed sensing and RSA encryption algorithm
CN116340757A (en) * 2023-04-25 2023-06-27 中国地震局地震研究所 Characteristic self-adaptive earthquake early-warning magnitude prediction method and system thereof

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103368578A (en) * 2013-07-01 2013-10-23 中国农业大学 Compressed-sensing-based signal sampling method for distributed wireless sensor network nodes
CN103745727A (en) * 2013-12-25 2014-04-23 南京邮电大学 Compressed sensing method of noise-containing voice signal
US20140195200A1 (en) * 2011-08-18 2014-07-10 University Of Antwerp Smart Data Sampling and Data Reconstruction
EP3206052A1 (en) * 2016-02-12 2017-08-16 CGG Services SAS Seismic data acquisition for compressive sensing reconstruction
US20190229842A1 (en) * 2016-10-10 2019-07-25 Shenzhen Super Data Link Technology Ltd. Signal sampling and recovery method and apparatus applicable to ovxdm system, and ovxdm system
CN111046737A (en) * 2019-11-14 2020-04-21 吉林大学 Efficient intelligent sensing acquisition method for microseism signal detection

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140195200A1 (en) * 2011-08-18 2014-07-10 University Of Antwerp Smart Data Sampling and Data Reconstruction
CN103368578A (en) * 2013-07-01 2013-10-23 中国农业大学 Compressed-sensing-based signal sampling method for distributed wireless sensor network nodes
CN103745727A (en) * 2013-12-25 2014-04-23 南京邮电大学 Compressed sensing method of noise-containing voice signal
EP3206052A1 (en) * 2016-02-12 2017-08-16 CGG Services SAS Seismic data acquisition for compressive sensing reconstruction
US20190229842A1 (en) * 2016-10-10 2019-07-25 Shenzhen Super Data Link Technology Ltd. Signal sampling and recovery method and apparatus applicable to ovxdm system, and ovxdm system
CN111046737A (en) * 2019-11-14 2020-04-21 吉林大学 Efficient intelligent sensing acquisition method for microseism signal detection

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
吕公河 等: "基于压缩感知的地震数据采集实践", 《石油物探》 *
文方青 等: "面向低信噪比的自适应压缩感知方法", 《物理学呗》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112965101A (en) * 2021-04-25 2021-06-15 福建省地震局应急指挥与宣教中心 Earthquake early warning information processing method
CN112965101B (en) * 2021-04-25 2024-03-08 福建省地震局应急指挥与宣教中心 Earthquake early warning information processing method
CN113573273A (en) * 2021-08-02 2021-10-29 尧米(重庆)智能科技有限公司 Wireless intelligent acquisition synchronization method and system based on Bluetooth transmission
CN113573273B (en) * 2021-08-02 2023-09-15 重庆优米工业自动化设备有限公司 Wireless intelligent acquisition synchronization method and system based on Bluetooth transmission
CN114647006A (en) * 2022-05-18 2022-06-21 东营市京川石油技术有限公司 Earthquake weak signal enhancement method based on multiple compressed sensing
CN114647006B (en) * 2022-05-18 2022-09-13 地大汇能（北京）科技有限公司 Earthquake weak signal enhancement method based on multiple compressed sensing
CN115996144A (en) * 2022-12-08 2023-04-21 西南石油大学 Fracturing data transmission method and system based on compressed sensing and RSA encryption algorithm
CN116340757A (en) * 2023-04-25 2023-06-27 中国地震局地震研究所 Characteristic self-adaptive earthquake early-warning magnitude prediction method and system thereof
CN116340757B (en) * 2023-04-25 2023-08-08 中国地震局地震研究所 Characteristic self-adaptive earthquake early-warning magnitude prediction method and system thereof

Also Published As

Publication number Publication date
CN111929724B (en) 2022-05-31

Similar Documents

Publication Publication Date Title
CN111929724B (en) Earthquake signal compressed sensing acquisition transmission method facing early warning
CN111458745B (en) Seismic signal sparse denoising method for early warning
KR101227443B1 (en) System and method for producing of parameters for earthquake early warning
CN113936142A (en) Rainfall approach forecasting method and device based on deep learning
CN103903261A (en) Spectrum image processing method based on partition compressed sensing
CN107561416A (en) A kind of local discharge signal acquisition system and method based on compressed sensing
CN110687587A (en) Medium-and-long-term earthquake early warning system based on multi-source remote sensing data and monitoring station data
WO2021109577A1 (en) Method and device for ultrasonic detection
CN104602297B (en) Radio sensor network monitoring method and system based on compressed sensing
CN116359617A (en) Lightning positioning method and system based on mixed base line
Wang et al. A method of real‐time tsunami detection using ensemble empirical mode decomposition
CN110263876A (en) A kind of patter of rain signal de-noising processing method and system
CN106027061A (en) Lamb wave compression sensing method based on adaptive observation matrix
CN115362367A (en) Compressed sensing for full matrix capture
CN104391040A (en) Phased array structure health monitoring Lamb wave compression and reconfiguration method
CN110166024A (en) A kind of faint acoustic marker signal enhancing method
CN105553894A (en) Blind source separation method based on parallel factor compressed sensing in sound vector array
Tang et al. Application of compressed sensing in the guided wave structural health monitoring of switch rails
CN104794571A (en) Method of acquiring mass battery data of storage power station
CN104243986A (en) Compressed video capture and reconstruction system based on data drive tensor subspace
Ghiurcau et al. A modified TESPAR algorithm for wildlife sound classification
Cao et al. Deformation data recovery based on compressed sensing in bridge structural health monitoring
CN111709279A (en) Algorithm for separating microseism noise mixed signals by utilizing SVD-EMD (singular value decomposition-empirical mode decomposition) algorithm
Hu et al. An energy-efficient Distributed Compressed Sensing architecture for Wireless Sensor Networks based on a distributed wavelet compression algorithm
Rusu et al. On the design of an acoustic based wildlife intruder detection system

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