IEEE Geoscience and Remote Sensing Magazine - September 2019 - 100

ELECTROMAGNETIC WAVE TOMOGRAPHY
In the 1990s, computer tomography techniques were applied
to radio wave penetration (called radio imaging methods) with
two types of reconstructions of the absolute or relative attenuation coefficients, both of which are based on the straight
ray assumption [20], [21], [100], [101]. Wu [102] offered
one practical algorithm to simulate the amplitudes of electromagnetic waves in the electrical model of a sandwich-like

coal bed. Liu and Yu [103] studied the attenuation characteristics of electromagnetic waves by numerical modeling in the
frequency domain. Yu et al. [104], [105] applied the finitedifference time-domain method to simulate the electromagnetic field in a coal seam. Liu et al. [106] proposed a precise
inversion method for tomography via an incident angle correction. Unfortunately, all of these electromagnetic tomography methods are based only on the amplitude attenuation
of radio waves, which has the inherent
disadvantage of poor resolution, espe13
12
11
10
9
cially in the case of objects parallel to
Surrounding Rock
the roadway. When there are groups
of structures nearly parallel to the
8
>100 m
roadway, the attenuation coefficient
Roadway
7
35 m
of the tomography would neither
Water Area
6
uniquely determine the position of
every structure nor distinguish them
Surrounding Rock
5
from each other. Also, when there are
1
2
3
4
several oblique linear anomalies close
(a)
to each other, it is difficult to distin30
20
10
7
5
3
2
0
guish each of them clearly according
Ω ⋅m
to the results of tomography with an
Station Number
attenuation coefficient.
Yu et al. [107] simulated the propagation of electromagnetic waves in
-30
-30
coal seams and found that phase char20
30
acteristics of the wave can improve
the resolution of its tomography; Liu
-40
-40
et al. [108] proposed a phase tomography technique for radio waves on the
basis of theoretical data. Yue [109] fur-50
-50
ther confirmed that phase information
can be accurately recognized (Figure 6)
even in the case of composite plates
parallel to roadway. On this basis, she
-60
-60
proposed a full-waveform curve observation scheme for electromagnetic toApparent
mography and developed a full-wave-70
-70
Resistivity
form probability inversion algorithm
on the basis of the Monte Carlo probability and statistics method.
-80
-80
The finite-difference time-domain
method
is used to simulate the ampli10
3
tude and phase characteristics of elec-90
-90
tromagnetic waves in coal seams when
they encounter two faults parallel to
the direction of the roadway. The size
-100
-100
of the two fault models is set to 20 m
0
5
10 15 20 25 30 35 40 45 50 55 60
× 2 m, and the distance between the
Relative Coordinate Values (m)
two faults along the roadway direc(b)
tion is set to 10 m. Figure 6(a) gives the
amplitude distribution of the electroFIGURE 5. Advance-detection results of MTEM. (a) The measuring points are arranged in a
ring shape, indicated by 1, 2, ..., 13. The arrows point to the normal directions of the transmit- magnetic wave, and Figure 6(b) shows
its phase distribution. The simulation
ting and receiving antennas. The gray dashed line represents a low-resistivity anomaly, which
results reveal that the two faults cancorresponds to water. (b) The top and bottom axes are measured point numbers and horizonnot be distinguished in the amplitude
tal distances, respectively, and the unit is meter for the ordinate and bottom axes.
10

Depth (m)

10
7

Depth (m)

100

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

SEPTEMBER 2019

IEEE Geoscience and Remote Sensing Magazine - September 2019

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - September 2019