Progress In Electromagnetics Research M, Vol.

94, 9–18, 2020

Coupling Analysis of Non-Parallel Transmission Lines Excited

by Ambient Wave Using a Time Domain Hybrid Method

Zhihong Ye1, * and Qingyuan Fang2

Abstract—A time domain hybrid method is presented to solve the coupling problem of non-parallel
transmission lines (NPTLs) excited by ambient wave efficiently, which consists of transmission line
(TL) equations, finite-difference time-domain (FDTD) method, and interpolation techniques. In this
method, NPTLs are firstly divided into multiple independent transmission line segments according to
the FDTD grids. Then TL equations are applied to build the coupling models of these TL segments,
which rely on the calculation precisions of per unit length (p.u.l) distribution parameters of NPTLs and
equivalent sources of TL equations. Thus, the p.u.l parameters of NPTLs are derived from empirical
formulas, and the equivalent sources are obtained by some linear interpolation schemes of electric fields
on the edges of FDTD grids. Finally, the difference scheme of FDTD is utilized to discretize the TL
equations to obtain the voltages and currents on NPTLs and terminal loads. The significant feature
of this hybrid method is embodied by its synchronous calculations of space electromagnetic fields and
transient responses on NPTLs in time domain. The accuracy of this presented method is testified by
the numerical simulations of plane wave coupling to NPTLs on the ground and in the shielded cavity
by comparing with FDTD-SPICE method and CST software.

1. INTRODUCTION

With the rapid development of wireless communications, the working frequencies and integration degrees
of electronic devices are becoming higher and higher. As a result, these devices are more susceptible
to space electromagnetic fields. Transmission lines (TLs) widely used in these devices are the main
paths of space electromagnetic fields coupling to the sensitive circuits of the lines, which may disturb or
damage the normal operations of these circuits. Limited to the positions of the circuits in these devices,
transmission lines cannot always keep parallel with other lines, which can be called as non-parallel
transmission lines (NPTLs). Therefore, to guarantee the safety of these devices, studying the coupling
problem of NPTLs excited by ambient wave is a significant way.
The preferred method used for the coupling analysis of NPTLs is full wave algorithms, such as
finite-difference time-domain (FDTD) method [1], method of moments (MOM) [2] and finite element
method (FEM) [3]. Unfortunately, the prominent feature of these methods is that the targets should
be meshed in fine grids when they contain local electrically small structures, which should occupy large
memories. In conclusion, full wave algorithms may not be suitable for the coupling analysis of electronic
devices obviously, because transmission lines are the important components of these devices, which are
fine structures.
Hence, studying field-to-line numerical methods is greatly necessary. Fortunately, several efficient
hybrid methods have been developed for decades. Among these methods, Baum-Liu-Tesche (BLT)
equation was first proposed in 1978 [4]. It aims at building the relationships of the voltages and
Received 13 May 2020, Accepted 18 June 2020, Scheduled 4 July 2020
* Corresponding author: Zhihong Ye ([email protected]).
1 School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065,

China. 2 Laboratory of Electromagnetic Environment Effects, Shijiazhuang Campus of Army Engineering University,

Shijiazhuang 050003, China.

10 Ye and Fang

currents at the nodes of transmission lines by scattering and transmission matrices, which are solved
by matrix operation to obtain the responses on these nodes. However, traditional BLT equation [5–
8] is a frequency-domain method, which is not suitable for the coupling simulation when ambient
wave is a broadband signal. Although this method has been extended to time domain [9], it needs a
number of convolution operations, which should decrease the computation efficiency of this method.
Thus, to better deal with the coupling problems of broadband signals acting on TLs, efficient time
domain algorithms are necessary to be studied. Finite-difference time-domain simulation program with
integrated circuit emphasis (FDTD-SPICE) and FDTD-TL methods are the two classical numerical
methods in time domain. In FDTD-SPICE method [10–14], the excitation fields of TLs are computed
by the FDTD method firstly, and then the SPICE equivalent circuit model of TLs are established and
solved by SPICE software to obtain the transient responses on the loads of TLs. Unfortunately, it
needs a number of theoretical derivations, and the calculations of electromagnetic fields and transient
responses on the loads are obtained separately. The significant feature of FDTD-TL method [15–17]
compared with FDTD-SPICE method is that it can realize the co-calculations of space electromagnetic
fields and transient responses on TLs and loads. In this method, the space electromagnetic fields are
computed by the FDTD method, which are imported into the TL equations as distribution sources.
Then the TL equations are solved by the difference scheme of FDTD method to obtain the responses
on the lines. However, it is just suitable for parallel transmission lines currently.
Therefore, this paper presents an efficient time domain hybrid method based on FDTD-TL method,
which can be well applied to the coupling analysis of ambient wave to non-parallel transmission lines.
Compared with FDTD-TL method, this presented method is improved in two aspects: (1) the p.u.l
distribution parameters suitable for NPTLs are derived from empirical formulas; (2) the interpolation
schemes used for the calculations of equivalent sources of NPTLs are presented. In addition, this method
can realize the synchronous calculations of space electromagnetic field radiation and transient responses
on the NPTLs.

2. THEORY OF THE TIME DOMAIN HYBRID METHOD

To explain the theory of this presented method clearly, a typical model of NPTLs on the ground is
considered, as shown in Fig. 1. The ground is assumed as a perfect conductor (PEC) plane and located
on the xy plane of cartesian coordinate system. These transmission lines with different heights and
oblique angles (named as θ1 , θ2 , . . . , θN ) are parallel to the ground. The horizontal projections of these
lines are assumed to be the same. Here, the radiation of NPTLs under the illumination of ambient
wave can be ignored, because the distances between the NPTLs and the ground are electrically small
compared with the minimum wavelength of the wave generally.

Figure 1. Typical model of NPTLs on the ground excited by ambient wave.

Progress In Electromagnetics Research M, Vol. 94, 2020 11

In this method, establishing the corresponding transmission line equations for the coupling analysis
of NPTLs is the most important and initial step, which can be expressed as
∂ ∂
V (y, t) + L I (y, t) = VF (y, t) (1)
∂y ∂t
∂ ∂
I (y, t) + C V (y, t) = IF (y, t) (2)
∂y ∂t
where V(y, t) and I(y, t) are the voltage and current vectors of NPTLs, respectively. L and C represent
the per unit length (p.u.l) inductance and capacitance matrices of NPTLs, respectively. VF (y, t)
and IF (y, t) stand for the equivalent distribution voltage and current sources, respectively, which are
described as
∂
VF (y, t) = − ET (y, t) + EL (y, t) (3)
∂y
∂
IF (y, t) = −C ET (y, t) (4)
∂t
ET (y, t) and EL (y, t) are obtained from the space electromagnetic fields, which are written as
 hi
[ET (y, t)]i = eex
z (x, y, z, t) dz (5)
0
[EL (y, t)]i = eex ex
li (x, y, hi , t) − eli (x, y, 0, t) (6)
where i stands for the i-th line, hi the height of the i-th line to the ground, and li the direction of the
i-th line. eex ex
l and ez are the incident electric fields along and perpendicular to the NPTLs, respectively.
The NPTLs can be removed when calculating the electromagnetic fields around them, because the
equivalent sources of TL equations are not relevant to the scattering fields of NPTLs.
Obviously, it can be observed from the TL equations that the precision of TL equations is
determined by the accurate calculations of p.u.l distribution parameters and equivalent sources of
NPTLs, which will be introduced in detail as follows.

2.1. Calculation of Per Unit Length Parameters for NPTLs

First, the NPTLs are divided into NL segments according to the FDTD grids, as shown in Fig. 2.

Figure 2. The division of NPTLs by FDTD grids.

The empirical formulas of inductance parameter matrix for parallel TLs can be found from [18],
which are expressed as
Lii = μ0 ln (2hi /ri ) /2π (7)
 
Lij = μ0 ln 1 + 4hi hj /d2ij /4π (8)
12 Ye and Fang

where i and j stand for the i-th and j-th transmission lines, respectively. Lii and Lij represent the self
and mutual inductances of the TLs, respectively. ri is the radius of the i-th line. hi , hj , and dij are the
heights of the i-th and j-th lines and the distance between the two lines, respectively.
Because the self inductance is independent of the distance between the lines, Equation (7) is still
suitable for the calculation of self inductances of NPTLs. On the other hand, mutual inductances are
closely related to the distances between the lines, and as a result, Equation (8) is not applicable to
the calculations of mutual inductances of NPTLs, because the space distances of NPTLs are changed
continually. Under this circumstance, to obtain the mutual inductances of each segment of NPTLs, it
should be further divided into multiple small elements, which can be seen as approximately parallel
transmission lines, as shown in Fig. 3. The number of elements is set as M , and the distances between
these elements are assumed as d1 , d2 , . . . , dM . Then the mutual inductances of each segment of NPTLs
can be obtained from the sum average of the mutual inductances of these elements, which can be
expressed as
k=M
   
Lij = μ0 ln 1 + 4hi hj /d2k /4π /M (9)
k=1

Then the capacitance parameter matrix of NPTLs can be calculated by C = μ0 ε0 L−1 .

Figure 3. The calculation of mutual inductance of NPTLs.

2.2. Calculation of Equivalent Distribution Sources for NPTLs

Because the distances between these lines of NPTLs are changed, the lines cannot be located on the
edges of FDTD grids. Therefore, the incident electric fields needed by the equivalent distribution sources
of NPTLs cannot be obtained from the electric fields on FDTD girds directly, which must be calculated
by interpolation schemes.
First, to get the values of EL (y, t) in distribution voltage source, the electric fields along the NPTLs
should be ascertained. To explain this solution process clearly, we take one line of NPTLs as an example,
and it is assumed that the starting and ending points of the line are on the edges of FDTD grids, as
shown in Fig. 4. Then the center point of each segment of this line is on the central plane of FDTD
gird where the point is located. The electric field along the line can be expressed by the vector product
form as E · el = Ex · ax ex + Ey · ay ey , where ex and ey stand for the unit vectors of x and y axes,
respectively. ax and ay represent the decomposed proportions of electric field along the line in x and
y axes, respectively, which are written as ax = sin θ and ay = cos θ, where θ is the angle between the
line and horizontal direction. Ex and Ey are the electric field components at the center point of each
segment, which should be obtained by the interpolation of four adjacent electric fields on the FDTD
grids. Taking Ey as an example, the interpolation scheme can be expressed as
Ey = (1 − η) [ξEy1 + (1 − ξ) Ey3 ] + η [ξEy2 + (1 − ξ) Ey4 ] (10)
where Ey1 , Ey2 , Ey3 , and Ey4 are the electric fields on FDTD grids. η and ξ are the scales of center
point in the FDTD grid along x and z directions.
Second, the values of ET (y, t) in distribution voltage and current sources must be confirmed, which
are contributed by the integration of electric fields Ezinc perpendicular to NPTLs at the starting and
Progress In Electromagnetics Research M, Vol. 94, 2020 13

Figure 4. Interpolation scheme for the electric fields at the center point of each segment.

Figure 5. Interpolation scheme for the electric fields at starting and ending points of each segment.

ending points of each segment of NPTLs, as shown in Fig. 5. Obviously, they can be interpolated from
two adjacent electric fields Ez on FDTD grids, which can be expressed as
Ezinc = αEz3 + (1 − α) Ez4 (11)

Note that the electric fields Ezinc near the lines should be interpolated from four adjacent electric fields
Ez on FDTD grids. The interpolation scheme can be described as

 [(0.5+β) [αEz1 + (1 − α) Ez2 ] + (0.5−β) [αEz3 + (1 − α) Ez4 ]] β < 0.5
Ezinc = (12)
[(1.5−β) [αEz1 + (1 − α) Ez2 ] + (β−0.5) [αEz5 + (1 − α) Ez6 ]] β ≥ 0.5
where α and β are the scales of starting or ending points in the FDTD grid along x and z directions.
After the TL equations are established, the difference scheme of FDTD is applied to solve the
TL equations to obtain the transient voltages and currents on the lines, which can reach the goal
14 Ye and Fang

of synchronous calculations of space electromagnetic fields and transient responses on NPTLs. The
iteration formulas of currents and voltages on the NPTLs are expressed as
⎛ ⎞
L (k) R (k) n− 1 1 Vn (k+1)−Vn (k)
− I 2 k+ −
n+ 21 1 R (k) L (k) −1 ⎜ ⎜ Δt 2 2 Δz ⎟
⎟
I k+ = + ⎜ ⎟ (13)
2 2 Δt ⎝ EnT (k + 1) − EnT (k) n 1 ⎠
− + EL k +
Δz 2
⎛ ⎞
n+ 21 1 n+ 12 1
⎜ I k+ −I k−
−1 ⎜ C (k) G (k) n (k) − 2 2 ⎟⎟
n+1 G (k) C (k) ⎜ − V ⎟
V (k) = + ⎜ Δt 2 Δz ⎟ (14)
2 Δt ⎜ ⎟
⎝ En+1 (k) − E n (k) ⎠
T
−C T
Δt
where k represents the positions of voltage and current nodes on the NPTLs. Δz and Δt are the space
and time steps used by the FDTD method, respectively.

3. NUMERICAL SIMULATION

Two cases about NPTLs on the PEC ground and in the shielded cavity are employed to test the accuracy
of this presented method by comparing with the FDTD-SPICE method and commercial software CST
using microwave studio.
The first case is the coupling analysis of NPTLs on the PEC ground excited by ambient wave,
as shown in Fig. 6. The dimension of the ground is 0.4 m × 1.0 m. The height and radius of
the three lines are 1.1 cm and 1 mm, respectively. If the original position of the ground is set as
(0, 0, 0), the positions of nodes #1, #2, #3, #4, #5, and #6 at the starting and ending ports of the
lines are (0.15 m, 0.1 m, 0.011 m), (0.19 m, 0.9 m, 0.011 m), (0.2 m, 0.1 m, 0.011 m), (0.2 m, 0.9 m, 0.011 m),
(0.23 m, 0.1 m, 0.011 m), and (0.21 m, 0.9 m, 0.011 m), respectively. The terminal loads of the lines are
Z1 = Z3 = Z5 = 50 Ω and Z2 = Z4 = Z6 = 100 Ω, respectively. A Gaussian pulse, expressed as
E0 exp[−4π(t − t0 )2 /τ 2 ], is perpendicular to the lines, where E0 = 1000 V/m, t0 = 1.6 ns, and τ = 2 ns.
To satisfy the stability condition of FDTD, the space and time steps selected by the hybrid method are
5 mm and 8.333 × 10−12 s, respectively.

Figure 6. Coupling model of MTLs on the PEC ground excited by ambient wave.

Figure 7 shows the voltage responses on the loads Z1 and Z6 computed by the hybrid method,
FDTD-SPICE, and CST. It can be seen clearly that the results of the three methods agree well to each
other. Generally, the feature selective validation (FSV) method [19, 20] is a preferred tool to access the
coincidence of two datasets, which is applied to compare the results obtained by this method and CST.
Here, the feature difference measure (FDM) analysis of FSV method is employed, which uses six levels
Progress In Electromagnetics Research M, Vol. 94, 2020 15

(a) (b)

Figure 7. Voltage responses on the loads of NPTLs computed by the three methods for the first case.
(a) Voltages on load Z1 . (b) Voltages on load Z6 .

(a) (b)

Figure 8. FDM analysis of the voltages for the first case. (a) Confidence histogram of FDM for the
voltages on Z1 . (b) Confidence histogram of FDM for the voltages on Z6 .

named as “EX”, “VG”, “G”, “F”, “P”, and “VP” to indicate the agreement degrees of the results. The
six levels represent “Excellent”, “Very Good”, “Good”, “Fair”, “Poor”, and “Very Poor”, respectively.
From Fig. 8 we can see that the good agreements of the oscillation periods of the voltages on Z1 and
Z6 obtained by the two methods are more than 94%.
The second case is the coupling analysis of NPTLs in the shielded cavity excited by ambient wave,
as shown in Fig. 9. The incident wave and the structures of NPTLs are the same with the first case.
The dimension of the cavity is Lc × Wc × Hc = 1 m × 0.3 m × 0.2 m. There are eight slots with size
ls × ws = 0.1 m × 0.01 m and distances ds = 0.15 m and dm = 0.02 m on the top surface of the cavity.
The terminal loads of NPTLs are also Z1 = Z3 = Z5 = 50 Ω and Z2 = Z4 = Z6 = 100 Ω.
In Fig. 10, the voltage responses on the loads Z1 and Z6 computed by the hybrid method, FDTD-
SPICE, and CST are observed to verify the accuracy of this presented method applied to the coupling
analysis of NPTLs in complex electromagnetic environment. Similarly, we can see that the results of
the three methods can keep good agreement.
16 Ye and Fang

Figure 9. Coupling model of NPTLs in the shielded cavity.

(a) (b)

Figure 10. Voltage responses on the loads of NPTLs computed by the three methods for the second
case. (a) Voltages on load Z1 . (b) Voltages on load Z6 .

In addition, it needs to explain that some values of the results computed by the hybrid method and
FDTD-SPICE comparing with that of CST for the two cases have certain errors, because the scattering
effects caused by the structures of NPTLs are ignored in this presented method and FDTD-SPICE
method.

4. CONCLUSION

Finite-difference time-domain method, transmission line equations, and some interpolation schemes are
combined to form a time domain hybrid method, which can solve the coupling problem of non-parallel
transmission lines excited by space electromagnetic fields with high precision. The contributions of this
method are represented in two aspects: (1) the calculation formulas suitable for the p.u.l distribution
parameters and equivalent sources of NPTLs are derived; (2) the synchronous computations of space
electromagnetic field radiation and transient responses on the NPTLs are realized.
Two cases about NPTLs, on the ground and in the shielded cavity respectively, excited by a plane
Progress In Electromagnetics Research M, Vol. 94, 2020 17

wave are studied to verify the accuracy of this presented method. To evaluate the performance of
this method, the results computed by the presented method are compared with that of FDTD-SPICE
method and commercial software CST, which are in a good degree of agreement. Next, this hybrid
method can be combined with the parallel technique to improve the efficiency of this method.

ACKNOWLEDGMENT

This work was supported by the National Natural Science Foundation of China (Grant No. 61701057),
and the Postdoctoral Science Foundation (Grant No. 2017M623400).

REFERENCES

1. Azizi, H., F. T. Belkacem, and D. Moussaoui, “Electromagnetic interference from shielding

effectiveness of a rectangular enclosure with apertures-circuital approach, FDTD and FIT
modeling,” Journal of Electromagnetic Waves and Applications, Vol. 28, 494–514, 2014.
2. Chen, Z. and S. Luo, “Generalization of the finite-difference-based time-domain methods using the
method of moments,” IEEE Transactions on Antennas and Propagation, Vol. 54, No. 9, 2515–2524,
2006.
3. Fu, W. N., Y. Zhao, S. L. Ho, and P. Zhou, “An electromagnetic field and electric circuit coupled
method for solid conductors in 3-D finite-element method,” IEEE Transactions on Magnetics,
Vol. 52, No. 3, 7401704, 2016.
4. Baum, C. E., T. K. Liu, and F. M. Tesche, “On the analysis of general multiconductor transmission-
line networks,” Interaction Notes, 350, 1978.
5. Du, J. K., S. M. Hwang, and J. W. Ahn, “Analysis of coupling effects to PCBs inside waveguide
using the modified BLT equation and full-wave analysis,” IEEE Transactions on Microwave Theory
and Techniques, Vol. 61, No. 10, 3514–3523, 2013.
6. Xie, L. and Y. Z. Lei, “Transient response of a multiconductor transmission line with nonlinear
terminations excited by an electric dipole,” IEEE Transactions on Electromagnetic Compatibility,
Vol. 51, No. 3, 805–810, 2009.
7. Yan, L. P., X. D. Zhang, and X. Zhao, “A fast and efficient analytical modeling approach for
external electromagnetic field coupling to transmission lines in a metallic enclosure,” IEEE Access,
Vol. 6, 50272–50277, 2018.
8. Nie, B. L., “Analysis of electromagnetic coupling to a shielded line based on extended BLT
equation,” Proceedings of Photonics & Electromagnetics Research Symposium, 2934–2937, 2019.
9. Tesche, F. M., “On the analysis of a transmission line with nonlinear terminations using the time-
dependent BLT equation,” IEEE Transactions on Electromagnetic Compatibility, Vol. 49, No. 2,
427–433, 2007.
10. Paul, C. R., “A SPICE model for multiconductor transmission lines excited by an incident
electromagnetic field,” IEEE Transactions on Electromagnetic Compatibility, Vol. 36, No. 4, 342–
354, 1994.
11. Erdin, I., A. Dounavis, and R. Achar, “A SPICE model for incident field coupling to lossy
multiconductor transmission lines,” IEEE Transactions on Electromagnetic Compatibility, Vol. 43,
No. 4, 485–494, 2001.
12. Xie, H., J. Wang, R. Fan, and Y. Liu, “SPICE models for radiated and conducted susceptibility
analyses of multiconductor shielded cables,” Progress In Electromagnetics Research, Vol. 103, 241–
257, 2010.
13. Xie, H. Y., J. G. Wang, and R. Y. Fan, “SPICE models for prediction of disturbances induced
by nonuniform fields on shielded cables,” IEEE Transactions on Electromagnetic Compatibility,
Vol. 53, No. 1, 185–192, 2011.
14. Chen, H. C., Y. P. Du, and M. Q. Yuan, “Lightning-induced voltages on a distribution line with
surge arresters using a hybrid FDTD-SPICE method,” IEEE Transactions on Power Delivery,
Vol. 33, No. 5, 2354–2363, 2018.
18 Ye and Fang

15. Ye, Z., X.-Z. Xiong, C. Liao, and Y. Li, “A hybrid method for electromagnetic coupling problems
of transmission lines in cavity based on FDTD method and transmission line equation,” Progress
In Electromagnetics Research M, Vol. 42, 85–93, 2015.
16. Ye, Z. H., J. J. Zhou, and D. Gou, “Coupling analysis of ambient wave to the shielded cavity with
penetrated wire using a time domain hybrid method,” Microwave and Optical Technology Letters,
Vol. 61, No. 11, 2551–2556, 2019.
17. Ye, Z. H., C. Liao, X. Z. Xiong, and M. Zhang, “A hybrid method combining the novel TD-
SC technique and FDTD method for the EMI analysis of transmission line network,” IEEE
Transactions on Electromagnetic Compatibility, Vol. 59, No. 4, 1211–1217, 2017.
18. Tesche, F. M., M. V. Ianoz, and T. Karlsson, EMC: Analysis Methods and Computational Models,
John Wiley & Sons, New York, NY, USA, 1997.
19. Duffy, A. P., A. J. M. Martin, A. Orlandi, G. Antonini, T. M. Benson, and M. S. Woolfson, “Feature
Selective Validation (FSV) for validation of Computational Electromagnetics (CEM). Part I — The
FSV method,” IEEE Transactions on Electromagnetic Compatibility, Vol. 48, No. 3, 449–459, 2006.
20. Orlandi, A., A. P. Duffy, B. Archambeault, G. Antonini, D. E. Coleby, and S. Connor, “Feature
Selective Validation (FSV) for validation of Computational Electromagnetics (CEM). Part II —
Assessment of FSV performance,” IEEE Transactions on Electromagnetic Compatibility, Vol. 48,
No. 3, 460–467, Aug. 2006.