CN107656272B - A kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm - Google Patents
A kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm Download PDFInfo
- Publication number
- CN107656272B CN107656272B CN201710872973.8A CN201710872973A CN107656272B CN 107656272 B CN107656272 B CN 107656272B CN 201710872973 A CN201710872973 A CN 201710872973A CN 107656272 B CN107656272 B CN 107656272B
- Authority
- CN
- China
- Prior art keywords
- time
- imaging
- ultra
- wideband radar
- field
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/885—Radar or analogous systems specially adapted for specific applications for ground probing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/887—Radar or analogous systems specially adapted for specific applications for detection of concealed objects, e.g. contraband or weapons
- G01S13/888—Radar or analogous systems specially adapted for specific applications for detection of concealed objects, e.g. contraband or weapons through wall detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/12—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation operating with electromagnetic waves
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Electromagnetism (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Life Sciences & Earth Sciences (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
A kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm is related to Computational electromagnetics and wideband radar detection.It is applicable not only to the buried target of ULTRA-WIDEBAND RADAR and hidden media area is reconnoitred and data acquisition imaging, be further widely applied to the particular problems such as geological prospecting, road prospecting and through-wall imaging.The experimental signal detected using three-dimensional high-order time-domain finite difference and ULTRA-WIDEBAND RADAR, the reverse field distribution for the time domain observation signal that the positive field distribution and ULTRA-WIDEBAND RADAR for obtaining the time domain excitation signal that ULTRA-WIDEBAND RADAR receives receive.Under the requirement for meeting reverse-time migration image-forming condition, theoretically inclination angle is overcome to limit, and can realize accurately image.The numerical simulation of reverse-time migration Imaging Simulation is carried out to target area using radar signal, the dynamic information of object to be measured can effectively be obtained, compared to traditional magnetography algorithm imaging resolution advantage with higher, take a firm foundation to obtain identification and the parametric inversion of object to be measured.
Description
Technical field
The present invention relates to Computational electromagnetics and wideband radar to detect, more particularly, to a kind of electricity under higher-Order Time-Domain algorithm
Magnetic wave three-dimensional reverse-time migration imaging method.
Background technique
In terms of wide spectrum electromagnetic technique, ULTRA-WIDEBAND RADAR method can effectively estimate current underground medium distribution.It is super
The working principle of wideband radar is to emit a broadband electromagnetic signal to target area using an antenna or mutiple antennas combination,
Another antenna or mutiple antennas combination receive interface reflection and the scattered signal of the object to be measured in target area.Electromagnetic pulse
When signal is propagated in object to be measured, the inevitable particular parameter with medium of its specific path, field strength and waveform profiles and
Its geometry state and acute variation.Although echo electromagnetic signal obtains the data such as its stroke, amplitude and phase based on the received,
It can simply infer the approximate location of medium, but cannot effectively determine its basic configuration and structure.In order to preferably be visited to imaging
Survey technology development, current maximum hot spot are that probing into for buried target and hidden medium is carried out using ULTRA-WIDEBAND RADAR, due to
Ultra wide band has the spectrum information of rich content, and spatial and temporal resolution also can be promoted preferably, this is to underground medium construction, underground mesh
Mark and hidden medium field are oriented to better researching value.
By the propagation characteristic of Electromagnetic Wave phenomenon, the scientist and researcher of geological research carry out numerical value using computer
Analogue simulation.In order to realize the ULTRA-WIDEBAND RADAR detection of target area and reconnoitre, more more options are maximally efficient using Computational electromagnetics
Time-domain calculation method, i.e. Finite-Difference Time-Domain Method (FDTD, Finite-Difference Time-Domain).In time-domain
It is discrete with direct differential is carried out to Maxwell equation group in spatial domain, and effectively simulate electromagnetic wave propagation process, this method
It is widely applied among the research such as microwave circuit, antenna technology and Electromagnetic Scattering of Target already.During target area is reconnoitred,
ULTRA-WIDEBAND RADAR and ground surface have certain height, need to carry out board design using layered medium.Due to computer resource
Finiteness, ULTRA-WIDEBAND RADAR to unbounded space emit electromagnetic pulse signal, need using completely permutation (PML,
Perfectly Matched Layer) technology carries out truncation, realize the unbounded processing in electromagnetic field wave problem.
Since ultra wideband radar system can realize the function of Ground Penetrating Radar, the requirement of Ground Penetrating Radar is it needs to be determined that ground surface
Object to be measured existing for lower section is easy to understand position and the general shape of underground object to be measured.In terms of imaging algorithm, using
Seismic wave emulation aspect reverse-time migration (RTM, Reverse Time Migration) imaging method the most mature.This RTM
Technology uses round trip wave process, can go out electromagnetic wave propagation process with accurate simulation using FDTD method, to multiple reflections
Region is precisely imaged, and effectively determines the specific location of object to be measured.Since RTM imaging method needs to calculate transmitting at the same time
The wave field of the forward-propagating in source and the synchronization in the reverse communication process in reception source is superimposed, especially three-dimensional buried target
And the problem of hidden media area, it has to face the difficulty to mass data.ULTRA-WIDEBAND RADAR is generally in search coverage
Residence time is very limited, with greater need for handling as early as possible.RTM imaging method need to carry out multicore acceleration processing.Meeting imaging requirements
Required precision under, Research Team has been proposed carrying out solving above-mentioned problem using high-order RTM imaging method.Utilize Taylor's grade
Number expansion is discrete to spatial domain progress higher difference, devises high order finite difference time domain method, and carry out high-order to PML technology
Optimization processing.In addition, accelerate using OpenMP concurrent technique the code of high-order RTM technology.
Summary of the invention
It is an object of the invention to be applicable not only to the buried target of ULTRA-WIDEBAND RADAR and hidden media area to reconnoitre sum number
The problems such as being imaged according to acquisition, being further widely applied to geological prospecting, road prospecting and through-wall imaging provides a kind of in higher-Order Time-Domain calculation
Electromagnetic wave three-dimensional reverse-time migration imaging method under method.
The invention adopts the following technical scheme:
1) bottom layer treatment is imaged as RTM using high-order FDTD, considers the general characteristic requirement of medium, lossy medium region
In Maxwell equation group can be expressed as following symmetric form:
Wherein σeFor conductivity tensor, σmFor equivalent magnetic loss tensor, to time local derviation using Second-Order Central Difference carry out from
It dissipates, obtains:
Since with first calculating magnetic field H, the rear electric field E that calculates is main processes of calculation, therefore sets the time step of magnetic field H in n+1/2
On, the time step of electric field E is so realized the time domain discrete difference scheme of formula (1) and (2), is obtained on n+1:
Hn+1/2=(μ Δ t-1+0.5σm)-1[(μΔt-1-0.5σm)Hn-1/2-CEn] (4)
En+1=(ε Δ t-1+0.5σe)-1[(εΔt-1-0.5σe)En+CHn+1/2] (5) wherein C is curl operatorSquare
Formation formula, is defined as:
It is an antisymmetric tensor, and formula (4) and (5) they are the time domain discrete matrix form of maxwell equation group, according to
Time step recursion, available spatially electromagnetic wave propagation process；
The operator of space differenceWithRespectively indicate the local derviation operator in three-dimensional space.For solving local derviation in space
Problem usually obtains first approximation derivative using Taylor series expansion.On the Yee cellular in FDTD algorithm, electric field strength E and
It is staggeredly discrete distribution that magnetic field strength H is presented on node.Therefore, by taking the direction x as an example, the expansion of Taylor series are as follows:
Wherein n is taken as positive even numbers.With the value for taking different n, available higher difference coefficient matrix equation are as follows:
At this moment, it as long as choosing different n, can obtain about three-dimensional high-order FDTD iteration coefficient, when the essence of space difference
It spends order and expands to ten quadravalences from second order, the spatial sampling point of Electromagnetic Simulation simulated domain can be allowed to need 12 points from every wavelength
3 points of every wavelength are reduced to, at this moment, required for capable of extremely efficient reducing on zoning using the FDTD of higher-order
Number of grid.
2) it is truncated using high-order PML as boundary, the specific method being truncated using high-order PML as boundary can are as follows:
Due to the finiteness of computer memory space, ULTRA-WIDEBAND RADAR emits electromagnetic pulse signal to unbounded space, needs using specific
Boundary condition adjust, three-dimensional high-order FDTD needs just to be able to maintain stable electro-magnetic wave absorption with the PML technology of same rank, no
Complete space truncation can cause very big numerical error, and meeting severe exacerbation is originally true after certain analog simulation
Calculated result.
For the boundary of layered medium, need to handle the axial continuation of progress on the boundary PML, the institute of difference approximation on boundary
With order based on the order of three-dimensional high-order FDTD, with ExFor, expression is as follows:
Wherein ψexzAnd ψexyBeing is recursive convolution item, can be respectively indicated are as follows:
Wherein aeyAnd beyIt can be given respectively according to the parameter presets of PML, it, can be in formula (10), (11) and (12)
See local derviation operator thereinWithFor three-dimensional high-order FDTD algorithm, it is approximate that dependent variable also does identical higher derivative
Difference processing, the unnecessary calculating error that can be effectively prevented from layering interfaces.
In order to further increase scientific research progress and Efficiency, the operational speed of a computer more needs further to be promoted.Using
The concurrent technique of computer, sharpest edges are that multiple central processing unit synthetic operations solve the same problem.In RTM imaging process
In, a large amount of calculating times all concentrate in the simulation of three-dimensional high-order FDTD.In Visual Studio environment, it can install
At this moment Intel Fortran compiler can be introduced directly into OpenMP and further speed up processing.
3) specific steps of RTM imaging algorithm, according to step 1) and 2), the main time domain for completing RTM imaging algorithm changes
For process, in order to complete ULTRA-WIDEBAND RADAR imaging moiety, each station antenna of ultra wideband radar system is in turn to buried target and hidden
Media area carries out electromagnetic pulse signal transmitting, and is received by other antennas in ultra wideband radar system, detecting
Signal be sent to RTM imaging workbench.
Electromagnetic wave three-dimensional reverse-time migration imaging method of the present invention under higher-Order Time-Domain algorithm, comprising the following steps:
(1) specific location of each station antenna on ULTRA-WIDEBAND RADAR is determined, and to the time domain excitation that ULTRA-WIDEBAND RADAR receives
Signal and time domain observation signal are handled one by one；
(2) initial model for constructing imaging simulation, determines the position of prime area, to each medium parameter of simulated environment into
Row pre-sets；
(3) positive three-dimensional high-order FDTD analog simulation is carried out to the time domain excitation signal that ULTRA-WIDEBAND RADAR receives, is
Inverse time reduction is carried out to the magnetic distribution, being distributed or with regular hour sample rate for pre-stored boundary field need to be carried out
Carry out stored boundary field distribution, until the time progressively increases to maximum imaging time from starting；
(4) time domain that the simulated environment to imaging initial model and parameter reset, and ULTRA-WIDEBAND RADAR received
Observation signal carries out inverse time processing, and carries out three-dimensional high-order FDTD analog simulation, and the time is gradually reduced to from maximum imaging time
Until initial time；
(5) in the operational process of step (4), to pre-stored boundary field be distributed carry out inverse time input, or to
Regular hour sample rate carrys out pre-stored boundary field distribution and carries out inverse time input after spline interpolation, realizes normal field simulation process
When magnetic distribution inverse time reduction, until the time is also gradually reduced to initial time from maximum imaging time；
(6) by step (4) and step (5), respectively obtain what the time domain observation signal that ULTRA-WIDEBAND RADAR receives generated
The positive field distribution that the time domain excitation signal that reverse field distribution and ULTRA-WIDEBAND RADAR receive generates, reverse field distribution and normal field
Distribution can satisfy RTM image-forming condition, and electric field RTM imaging results and magnetic field RTM imaging results can use formulae express respectively
It is as follows:
Wherein r indicates the position vector on imaging region, and t indicates at the time of point, T in RTM imaging simulationmaxIndicate RTM at
As the maximum moment point in emulation, ETThe time domain excitation signal that expression ULTRA-WIDEBAND RADAR receives corresponds to the electric field point of normal field
Cloth, HTThe time domain excitation signal that expression ULTRA-WIDEBAND RADAR receives corresponds to the Distribution of Magnetic Field of normal field, ERIndicate that ULTRA-WIDEBAND RADAR connects
The time domain observation signal received corresponds to the field distribution of reverse field, HRIndicate the time domain observation signal pair that ULTRA-WIDEBAND RADAR receives
Answer the Distribution of Magnetic Field of reverse field.
(7) general RTM imaging is related to the case where multistation antenna carries out rotation transmitting and rotation reception.At this point, according to super
Multistation of the wideband radar in practical operation receives and dispatches design requirement, repeats step (2)~(6), and by newly-generated imaging results
IEAnd IHThe imaging results I calculated respectively with the last timeEAnd IHIt is overlapped processing.
(8) due to the energy roundtrip between object structures and imaging object complicated and changeable in imaging region, RTM at
As result IEAnd IHVarious noise signals are inevitably present, these noise signals will lead to the erroneous judgement to object to be measured.In order to
This problem is solved, after each station antenna in ULTRA-WIDEBAND RADAR completes imaging, obtains last RTM imaging results IEAnd IH,
The filtering and denoising on three-dimensional information are carried out, the present invention can effectively improve using three-dimensional Laplce's filtering method
Imaging effect exports preferable image.
Bottom layer treatment is imaged as RTM using high-order FDTD in the present invention, is accelerated using OpenMP technology.Work as space parallax
The precision order divided expands to ten quadravalences from second order, and the spatial sampling point in analogue simulation region can be allowed to need 12 from every wavelength
Point is reduced to 3 points of every wavelength.At this moment, using the FDTD of higher-order space difference, it can extremely efficient reduce and calculate area
Required number of grid on domain.
In order to which analog electromagnetic field is in the process of unbounded space, the completely permutation (PML) of Convolution-type is used.Three-dimensional high-order
FDTD needs the PML of the Convolution-type in conjunction with same rank, could effectively keep stable electro-magnetic wave absorption, avoid in incomplete sky
Between truncation when can generate very big numerical error, and realize and can still obtain true meter after certain analog simulation
Calculate result.
The present invention theoretically overcomes inclination angle limitation, and accurately image may be implemented.Using radar signal to target
Area carries out the numerical simulation of RTM Imaging Simulation, can effectively obtain the dynamic information of object to be measured, compared to other traditional electromagnetism
Imaging algorithm imaging resolution advantage with higher, RTM imaging technique can be the identification and parametric inversion of acquisition object to be measured
It lays a solid foundation.
In ultra wideband radar system, there is the specific location of fixed number of antennas and corresponding each station antenna.In order to mention
The effect and precision of height imaging, are typically chosen rotation transceiver mode.This mode refers to each station on ultra wideband radar system
Antenna emits electromagnetic pulse signal into space one by one, and reflected echo electromagnetism in remaining day line options reception space
Signal.The present invention can emit the various electromagnetic pulse signals of any selection by ultra wideband radar system to carry out analogue simulation,
Obtain the imaging results of target area.
The sharpest edges of RTM imaging algorithm of the invention embody as follows respectively:
1) the precision order of space difference expands to ten quadravalences from second order, can allow the spatial sampling point in analogue simulation region
12 points are needed to be reduced to 3 points of every wavelength from every wavelength；
2) continuation processing is carried out to the boundary in delamination area using Convolution-type PML technology, is effectively prevented from layered boundary
Reflection；
3) according to the vorticity equation analogue simulation of Maxwell equation group, the round trip wave process of electric field strength and magnetic field strength,
Realize RTM imaging；
4) it uses OpenMP concurrent technique and multicore acceleration is carried out to RTM program code；
5) by boundary condition technology and interpolation technique, the calculator memory in RTM imaging process is greatly reduced；
6) by three-dimensional laplacian spectral radius filtering technique, the interference waveform of RTM imaging is eliminated, optimizes RTM imaging effect.
7) RTM imaging technique can lay a solid foundation for the identification and parametric inversion for obtaining object to be measured.
Detailed description of the invention
Fig. 1 is electromagnetic wave three-dimensional reverse-time migration imaging method flow chart of the invention；
Fig. 2 is the detection model figure of ultra wideband radar system of the invention on buried target and hidden media area；
Fig. 3 is the time domain normalized signal that a certain station ultra wideband radar system of the invention receives；
Fig. 4 is three-dimensional Laplce's filter operator structure chart of the invention；
Fig. 5 is the longitudinal profile RTM imaging effect figure of three-dimensional simulation model of the present invention；
Fig. 6 is the horizontal section RTM imaging effect figure of three-dimensional simulation model of the present invention；
Fig. 7 is the inverse time application condition figure of present invention three-dimensional RTM imaging.
Specific embodiment
The present invention is further detailed below in conjunction with specific accompanying drawings and embodiments.
A specific embodiment of the invention is as follows:
Proposed by the present invention is a kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm, design
For the flow diagram of scheme referring to Fig. 1, block diagram information difference is as follows:
(1) acquire experimental signal: if to realize reverse-time migration imaging method, acquire experimental signal process be can not or
Scarce.It needs to handle each station time-domain signal of ultra wideband radar system.
(2) the time domain excitation signal of ULTRA-WIDEBAND RADAR: in collected experimental signal, the signal that will launch is needed
It extracts, and carries out corresponding normal field analog portion.
(3) it constructs initial model: for normal field analog portion, being related to the process of numerical experiment, the mesh of subsurface investigation
It is also unknown for marking, so building initial model by known situation.
(4) high-order FDTD is simulated: by high-order FDTD method, obtaining the information of each time step at all normal fields.
(5) whether boundary field partially stores: if so, carrying out part stored boundary field value；If it is not, then direct stored boundary
Enter next judgement after.
(6) part stored boundary field value: in view of the requirement of sampling theorem, boundary field information can be by part storage come also
The field simulation of original high-order FDTD method, and the magnanimity calculator memory being further reduced in normal field simulation process.
(7) whether time step terminates: for the electromagnetic calculation of time domain, time step is limited.If it is not, not reaching then
It is required that maximum time walk nmax, then it is still to carry out normal field simulation.If so, needing the next process into algorithm.
(8) the time domain observation signal of ultra wideband radar system: in collected experimental signal, need will be by underground matchmaker
Reflected electromagnetic signal extracts in matter, and carries out corresponding reverse field analog portion.
(9) spline interpolation restores initial field distribution: since the information of normal field only retains the information of boundary field, needing to pass through
Spline interpolation carries out numerical value reduction.
(10) high-order FDTD is simulated: it is defeated that the initial model of the time domain observation signal of ULTRA-WIDEBAND RADAR in simulations carries out the inverse time
Enter, realizes the analog simulation of high-order FDTD.
(11) the field distribution reduction of wave source excitation: by the inverse time calculating to field distribution, original forward direction is restored
Field distribution.
(12) image-forming condition: the positive field distribution of reverse field distribution and step (11) in step (10) directly constitutes
Image-forming condition, the RTM imaging results of available n-th time step.
(13) whether time step 1 terminates: judging whether time step is 1, if it is not, then needing to continue to be superimposed RTM imaging results；
If so, jumping out the circulation of RTM imaging moiety.
(14) export and be superimposed RTM imaging data: each station antenna requires to be overlapped place to the imaging results of RTM
Reason.
(15) it whether there is next station antenna: in ultra wideband radar system, it is understood that there may be multistation antenna emits in turn
Electromagnetic signal.If so, all steps for (1)~(14) that repeat the above steps；If it is not, then carrying out next step RTM imaging.
(16) it three-dimensional Laplce's filtering processing: for the information of three-dimensional RTM imaging, needs to be sharpened processing, makes RTM
Imaging results are further optimized.
(17) it exports image: being realized by drawing software, the RTM image under the medium of underground is showed,
Determine corresponding target position and general shape.
In ultra wideband radar system, there is the specific location of fixed number of antennas and corresponding each station antenna.In order to mention
The effect and precision of height imaging, are typically chosen rotation transceiver mode.This mode refers to each station on ultra wideband radar system
Antenna emits electromagnetic pulse signal into space one by one, and reflected echo electromagnetism in remaining day line options reception space
Signal.The present invention can emit the various electromagnetic pulse signals of any selection by ultra wideband radar system to carry out analogue simulation,
Obtain the imaging results of target area.Referring to fig. 2, it is measured by the characteristic to underground medium, constructs initial simulation model,
Ultra wideband radar system is placed in the top of underground medium, the rock at the 0.5m of underground is detected, to each station antenna point
Not Fa She electromagnetic pulse signal, the wherein station electro-magnetic signal information being connected to, as shown in Figure 3.
In order to complete radar detection imaging moiety, using each station antenna on ultra wideband radar system in turn to buried target
And hidden media area emits electromagnetic pulse, and carries out battle array reception by other antennas in ultra wideband radar system, detecting
Signal be sent to RTM imaging workbench, carry out following steps:
1) it determines the specific location of each station antenna on ULTRA-WIDEBAND RADAR, and the time domain excitation that ULTRA-WIDEBAND RADAR receives is believed
Number and time domain observation signal handled one by one, wherein a certain time domain normalized signal that receives of ULTRA-WIDEBAND RADAR of standing, such as Fig. 3
It is shown；
2) initial model for constructing imaging simulation, determines the position of prime area, to each medium parameter of simulated environment into
Row pre-sets；
3) positive three-dimensional high-order FDTD analog simulation is carried out to the time domain excitation signal that ULTRA-WIDEBAND RADAR receives, in order to
Inverse time reduction is carried out to the magnetic distribution, being distributed or with regular hour sample rate for pre-stored boundary field need to be carried out
Stored boundary field distribution, until the time progressively increases to maximum imaging time from starting；
4) time domain that the simulated environment to imaging initial model and parameter reset, and ULTRA-WIDEBAND RADAR received is seen
It surveys signal and carries out inverse time processing, and carry out three-dimensional high-order FDTD analog simulation, until the time is gradually reduced to 1 from maximum value；
5) in the operational process of step 4), inverse time input is carried out to being distributed for pre-stored boundary field, or to one
Fixed Temporal sampling carrys out pre-stored boundary field distribution and carries out inverse time input after spline interpolation, when realizing normal field simulation process
The inverse time reduction process of magnetic distribution, until the time is also gradually reduced to initial time from maximum imaging time；
6) by step 4) and 5), the reverse field point that the time domain observation signal that ULTRA-WIDEBAND RADAR receives generates is respectively obtained
The positive field distribution that time domain excitation signal that cloth and ULTRA-WIDEBAND RADAR receive generates, reverse field distribution and positive field distribution can be with
Meet RTM image-forming condition, electric field RTM imaging results and magnetic field RTM imaging results respectively can be as follows with formulae express:
Wherein r indicates the position vector on imaging region, and t indicates at the time of point, T in RTM imaging simulationmaxIndicate RTM at
As the maximum moment point in emulation, ETThe time domain excitation signal that expression ULTRA-WIDEBAND RADAR receives corresponds to the electric field point of normal field
Cloth, HTThe time domain excitation signal that expression ULTRA-WIDEBAND RADAR receives corresponds to the Distribution of Magnetic Field of normal field, ERIndicate that ULTRA-WIDEBAND RADAR connects
The time domain observation signal received corresponds to the field distribution of reverse field, HRIndicate the time domain observation signal pair that ULTRA-WIDEBAND RADAR receives
Answer the Distribution of Magnetic Field of reverse field.
7) general RTM imaging is related to the case where multistation antenna carries out rotation transmitting and rotation reception.At this point, according to ultra-wide
Design requirement is received and dispatched with multistation of the radar in practical operation, repeats step 2)~6), and by newly-generated imaging results IEAnd IH
The imaging results I calculated respectively with the last timeEAnd IHIt is overlapped processing.
8) due to the energy roundtrip between object structures and imaging object complicated and changeable in imaging region, RTM at
As result IEAnd IHVarious noise signals are inevitably present, these noise signals will lead to the erroneous judgement of object to be measured.To understand
Certainly this problem obtains last RTM imaging results I after each station antenna in ULTRA-WIDEBAND RADAR completes imagingEAnd IH, be into
Filtering and denoising on row three-dimensional information, the present invention are used using three-dimensional Laplce's filtering method, Fig. 4 present invention
Three-dimensional Laplce's filter operator structure chart, it can effectively improve imaging effect, export preferable image, such as Fig. 5
With shown in Fig. 6.
From the tool that can correctly reflect ULTRA-WIDEBAND RADAR detection object under test on buried target and hidden media area in Fig. 5
Body position situation can clearly know the substantial transverse size of object at this location from Fig. 6.In order to verify spline interpolation
To the sample rate situation of imaging, it can recognize from Fig. 7, when interpolation sampling rate is higher than 3, the deviation of imaging be can reduce
Within 1 percent, validity and feasibility of the present invention on interpolation technique are demonstrated.Using radar signal to target area into
The numerical simulation of row RTM Imaging Simulation can effectively obtain the dynamic information of object to be measured, compared to other traditional magnetographies
Algorithm imaging resolution advantage with higher.RTM imaging technique can be effectively anti-for the identification and parameter of acquisition object to be measured
It drills and lays a solid foundation.
Claims (3)
1. a kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm, it is characterised in that including following step
It is rapid:
(1) specific location of each station antenna on ULTRA-WIDEBAND RADAR is determined, and to the time domain excitation signal that ULTRA-WIDEBAND RADAR receives
It is handled one by one with time domain observation signal；
(2) initial model for constructing imaging simulation, determines the position of prime area, carries out to each medium parameter of simulated environment pre-
Setting；
(3) positive three-dimensional high-order FDTD analog simulation is carried out to the time domain excitation signal that ULTRA-WIDEBAND RADAR receives, to electromagnetism
Field distribution carries out inverse time reduction, carry out pre-stored boundary field be distributed or with Temporal sampling stored boundary field distribution, when
Between progressively increase to maximum imaging time from starting until；
(4) simulated environment and parameter of imaging initial model are reset, the time domain observation letter that ULTRA-WIDEBAND RADAR is received
Number inverse time processing is carried out, and three-dimensional high-order FDTD analog simulation is carried out, when the time is gradually reduced to starting from maximum imaging time
Between until；
(5) in the operational process of step (4), inverse time input is carried out to being distributed for pre-stored boundary field, or to the time
The inverse time inputs after sample rate stored boundary field distribution carries out spline interpolation, and magnetic distribution is inverse when realizing normal field simulation process
When restore, until the time is also gradually reduced to initial time from maximum imaging time；
(6) by step (4) and step (5), the reverse of the time domain observation signal generation that ULTRA-WIDEBAND RADAR receives is respectively obtained
The positive field distribution that the time domain excitation signal that field distribution and ULTRA-WIDEBAND RADAR receive generates, reverse field distribution and positive field distribution
Meet RTM image-forming condition, electric field RTM imaging results and magnetic field RTM imaging results are as follows with formulae express respectively:
Wherein r indicates the position vector on imaging region, and t indicates at the time of point, T in RTM imaging simulationmaxIndicate that RTM imaging is imitative
Maximum moment point in very, ETThe time domain excitation signal that expression ULTRA-WIDEBAND RADAR receives corresponds to the field distribution of normal field, HTTable
Show that time domain excitation signal that ULTRA-WIDEBAND RADAR receives corresponds to the Distribution of Magnetic Field of normal field, ERIndicate what ULTRA-WIDEBAND RADAR received
Time domain observation signal corresponds to the field distribution of reverse field, HRIndicate that the time domain observation signal that ULTRA-WIDEBAND RADAR receives is corresponding reverse
The Distribution of Magnetic Field of field；
(7) RTM imaging is related to the case where multistation antenna carries out rotation transmitting and rotation reception, according to ULTRA-WIDEBAND RADAR in reality
Multistation when operation receives and dispatches design requirement, repeats step (2)~(6), and by newly-generated imaging results IEAnd IHRespectively with upper one
The imaging results I of secondary calculatingEAnd IHIt is overlapped processing；
(8) due to the energy roundtrip between object structures and imaging object complicated and changeable in imaging region, RTM imaging knot
Fruit IEAnd IHThere are noise signal, the noise signal leads to the erroneous judgement to object to be measured, when each station antenna in ULTRA-WIDEBAND RADAR
After completing imaging, last RTM imaging results I is obtainedEAnd IH, the filtering and denoising on three-dimensional information are carried out, using three-dimensional
Laplce's filtering method effectively improves imaging effect, exports image.
2. a kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm as described in claim 1, feature
It is that bottom layer treatment is imaged as RTM using high-order FDTD, is accelerated using OpenMP technology, when the accuracy order of space difference
Number expands to ten quadravalences from second order, and the spatial sampling point in analogue simulation region is allowed to need 12 points to be reduced to every wavelength from every wavelength
3 points can extremely efficient reduce the required number of grid on zoning using the FDTD of high order spatial difference.
3. a kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm as described in claim 1, feature
It is analog electromagnetic field in the process of unbounded space, using the completely permutation of Convolution-type, three-dimensional high-order FDTD needs to combine same
The PML of the Convolution-type of equal ranks, keeps electro-magnetic wave absorption, avoids generating numerical error in incomplete space truncation, and
It realizes and still obtains true calculated result after analog simulation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710872973.8A CN107656272B (en) | 2017-09-25 | 2017-09-25 | A kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710872973.8A CN107656272B (en) | 2017-09-25 | 2017-09-25 | A kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107656272A CN107656272A (en) | 2018-02-02 |
| CN107656272B true CN107656272B (en) | 2019-08-20 |
Family
ID=61131283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710872973.8A Active CN107656272B (en) | 2017-09-25 | 2017-09-25 | A kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107656272B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200116881A1 (en) * | 2018-10-12 | 2020-04-16 | Massachusetts Institute Of Technology | Heterogeneous subsurface imaging systems and methods |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11841265B2 (en) | 2018-10-12 | 2023-12-12 | The Trustees Of Boston University | Heterogeneous subsurface imaging systems and methods |
| CN109613490B (en) * | 2018-12-21 | 2022-12-06 | 中国科学院国家天文台 | Method for judging validity of lunar radar detection data horizon signal |
| US11106841B1 (en) * | 2019-04-29 | 2021-08-31 | X Development Llc | Physical device optimization with reduced memory footprint via time reversal at absorbing boundaries |
| CN111537996A (en) * | 2020-06-02 | 2020-08-14 | 西安石油大学 | Through-wall radar imaging method based on convolutional neural network |
| CN113686329B (en) * | 2021-08-27 | 2023-07-25 | 中国人民解放军国防科技大学 | Geomagnetic data-based vertical altitude potential field measurement method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101576622A (en) * | 2009-06-12 | 2009-11-11 | 成都理工大学 | Simulation method of ultra wideband electromagnetic wave |
| CN102269820A (en) * | 2010-06-01 | 2011-12-07 | 潜能恒信能源技术股份有限公司 | Three-dimensional seismic pre-stack reverse-time migration imaging method based on GPU (graphics processing unit) staggered grid with small memory capacity |
| CN101995573B (en) * | 2009-08-12 | 2012-09-26 | 中国科学院电子学研究所 | Target boundary imaging method for ultra-wide-band through-wall radar |
| US9329290B2 (en) * | 2010-07-28 | 2016-05-03 | Cggveritas Services Sa | 3-D harmonic-source reverse time migration systems and methods for seismic data analysis |
| CN105717539A (en) * | 2016-01-28 | 2016-06-29 | 中国地质大学（北京） | Multi GPU calculation based reverse time migration imaging method of 3D TTI medium |
| GB2490215B (en) * | 2011-04-12 | 2017-02-22 | Cggveritas Services Sa | Device and method for calculating 3D reverse time migration in tilted orthorhombic media |
| CN107102353A (en) * | 2017-05-08 | 2017-08-29 | 厦门大学 | Equations for elastic waves reverse-time migration imaging method based on High-order Difference Methods |
-
2017
- 2017-09-25 CN CN201710872973.8A patent/CN107656272B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101576622A (en) * | 2009-06-12 | 2009-11-11 | 成都理工大学 | Simulation method of ultra wideband electromagnetic wave |
| CN101995573B (en) * | 2009-08-12 | 2012-09-26 | 中国科学院电子学研究所 | Target boundary imaging method for ultra-wide-band through-wall radar |
| CN102269820A (en) * | 2010-06-01 | 2011-12-07 | 潜能恒信能源技术股份有限公司 | Three-dimensional seismic pre-stack reverse-time migration imaging method based on GPU (graphics processing unit) staggered grid with small memory capacity |
| US9329290B2 (en) * | 2010-07-28 | 2016-05-03 | Cggveritas Services Sa | 3-D harmonic-source reverse time migration systems and methods for seismic data analysis |
| GB2490215B (en) * | 2011-04-12 | 2017-02-22 | Cggveritas Services Sa | Device and method for calculating 3D reverse time migration in tilted orthorhombic media |
| CN105717539A (en) * | 2016-01-28 | 2016-06-29 | 中国地质大学（北京） | Multi GPU calculation based reverse time migration imaging method of 3D TTI medium |
| CN107102353A (en) * | 2017-05-08 | 2017-08-29 | 厦门大学 | Equations for elastic waves reverse-time migration imaging method based on High-order Difference Methods |
Non-Patent Citations (2)
| Title |
|---|
| Three-dimensional reverse-time migration applied to a MIMO GPR system for subsurface imaging;H. Liu et al.;《2016 16th International Conference on Ground Penetrating Radar (GPR)》;20160922;第1-4页 |
| 机载探地雷达数值模拟及逆时偏移成像;傅磊等;《地球物理学报》;20140531;第57卷(第5期);第1636-1646页 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200116881A1 (en) * | 2018-10-12 | 2020-04-16 | Massachusetts Institute Of Technology | Heterogeneous subsurface imaging systems and methods |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107656272A (en) | 2018-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107656272B (en) | A kind of electromagnetic wave three-dimensional reverse-time migration imaging method under higher-Order Time-Domain algorithm | |
| Ernst et al. | Application of a new 2D time-domain full-waveform inversion scheme to crosshole radar data | |
| CN104471443B (en) | Use the method for near field and far field ultralow frequency and very low frequencies Interferometric Synthetic Aperture Radar for subsurface imaging | |
| CN103698763B (en) | Based on the linear array SAR sparse formation method of hard-threshold orthogonal matching pursuit | |
| CN105158808B (en) | A kind of shallow sea transient electromagnetic air-sea detection and its means of interpretation | |
| CN102830401B (en) | Windowing weighted backward projection imaging method for ground penetrating radar | |
| Catapano et al. | GPR imaging via qualitative and quantitative approaches | |
| CN107037429A (en) | Linear array SAR three-D imaging methods based on thresholded gradient tracing algorithm | |
| Xiao et al. | Ground experiments of Chang’e-5 lunar regolith penetrating radar | |
| Aboudourib et al. | A processing framework for tree-root reconstruction using ground-penetrating radar under heterogeneous soil conditions | |
| CN105866772A (en) | Novel method for positioning metal object in human body based on microwave coherence imaging | |
| CN106199524B (en) | Far field Broadband RCS data acquisition and the compression method of denoising are tracked based on base | |
| CN103439707B (en) | Method for migration imaging of windowing distance of ground penetrating radar | |
| CN109061750A (en) | A kind of method and device of detection hot dry rock probing target area | |
| Wu et al. | A novel FDTD-based 3-D RTM imaging method for GPR working on dispersive medium | |
| Jia et al. | Time-domain inverse scattering method for cross-borehole radar imaging | |
| Tajdini et al. | Real-time modeling of forward-looking synthetic aperture ground penetrating radar scattering from rough terrain | |
| Hintz | SNR improvements in NIITEK ground-penetrating radar | |
| CN105891826A (en) | Airborne radar fast maximum posteriori imaging method | |
| Galli et al. | 3D imaging of buried dielectric targets with a tomographic microwave approach applied to GPR synthetic data | |
| CN113534140B (en) | Ground penetrating radar three-dimensional imaging method based on wave field cross correlation | |
| Tajdini et al. | Multiview synthetic aperture ground-penetrating radar detection in rough terrain environment: A real-time 3-d forward model | |
| CN106405509B (en) | The piecemeal processing method of space-time adaptive signal | |
| Duman et al. | 3D imaging for ground-penetrating radars via dictionarydimension reduction | |
| Yi et al. | Fusion before imaging method for heterogeneous borehole radar subsurface surveys |
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 |