1536-1225 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/LAWP.2015.2471079, IEEE Antennas and Wireless Propagation Letters
> <
Abstract—The parabolic equation (PE) method is employed to
solve the Loran-C additional secondary factors (ASFs) over
irregular terrain. Based on the split-step Fourier transform
(SSFT) algorithm, the method has been proven to be numerically
efficient. The ASF results are compared to those of the integral
equation (IE) method and the finite-difference time-domain
(FDTD) method. Very good agreements are observed. The
computational time of the PE method is several orders less than
that of the other two. The memory requirement is similar to the IE,
and less than the FDTD method.
Index Terms—Parabolic Equation (PE), additional secondary
factors (ASFs), irregular terrain, split-step Fourier transform
(SSFT).
I. INTRODUCTION
GROUND wave propagating over complex paths with
varying electric parameters and rough terrains causes an
extra delay named additional secondary factor (ASF) [1]. It is a
significant factor ensuring the precision of longwave
navigation system and timing service. A small deviation of
ASF may result in large errors up to several kilometers [2].
Over the years, researchers have developed simplified
simulation models to calculate the ASF with different
assumptions and approximations. The three most commonly
used ones are the smooth-earth (SE), smooth-earth mixed-path
(SEMP), and irregular-earth mixed-path (IEMP) models where
the former two are very mature during the past years [3].
Formulas and calculation methods are built up based on these
models. For example, the flat-earth formula and residue Series
formula are developed based on the SE model with the
assumption that the propagation path is smooth, and the electric
parameters are homogeneous [4], [5]. The Wait integral
equation, Millington empirical formula [5], [6], and wave mode
conversion method are based on the SEMP model [7]. The
algorithms for IEMP model are usually numerical ones, such as
the integral equation (IE) [8], [9] and finite-difference
time-domain (FDTD) method [10], [11]. The IE method is
based on the assumption that the impacts from backscattering
waves are very small and can be neglected. As expected, it has a
This work was supported in part by the National Natural Science Foundation
of China (No. 61271091 and No. 61401261) and the Scientific Research
Program Funded by Shaanxi Provincial Education Department (Program No.
14JK1513 and No. 14JK1084).
Corresponding Author is X.L. Xi, e-mail: xixiaoli@xaut.edu.cn.
D. D. Wang, X. L. Xi, Y. R. Pu, and J. F. Liu are with Faculty of Automation
and Information Engineering, Xi’an University of Technology, Xi’an, 710048,
China.
L. L. Zhou is with College of Electrical and Information Engineering,
Shaanxi University of Science and Technology, Xi’an, 710021, China.
large error for steeper terrain and choppy electric parameters
and is hard to be used for a three-dimension (3D) case because
of the vulnerability of the algorithm itself. The FDTD method
has been proven to be most the precise one for complex paths
with irregular terrain, but the computational expenditures of
memory and time are huge for the large area prediction.
The parabolic equation (PE) is an approximation of the
Helmholtz equation assuming that the energy propagates in the
paraxial direction [12]. Numerical methods for solving the PE
can categorize into the finite-difference (FD) and the split-step
Fourier-transform (SSFT) algorithms [12]-[16]. The FD
algorithm is capable for complex terrains but needs fine
sampling of the range, leading to huge computational
expenditures [17]. The SSFT algorithm allows for a fairly large
range step, but the boundary condition must be enforced to flat
[12], [13]. The main issue with the PE method is its poor
accuracy for short propagation ranges with large angles due to
its paraxial approximation nature [18]. Therefore, the initial
field injection needs special attention. In [19]- [22], the PE
method has been successfully used to predict the path loss of
ground wave propagation in the HF/VHF band, where the
initial fields are injected using the near-field/far-field
transformation of the antenna pattern.
In this letter, we employ the PE method for Loran-C ASF
prediction over irregular terrain. We propose a hybrid solution
that uses analytical equations to calculate the initial field for the
PE method. The field distribution at a cross-sectional plane
with a certain distance to the radiation source is calculated
analytically, where the Earth surface must be flat between the
source and the plane. The PE method then takes the field on the
cross-sectional plane as input and continues to solve the rest
propagation path. Using the SSFT algorithm, the PE method
has higher efficiency and accuracy when compared with the IE
and FDTD method. This letter is organized as follows: Section
II outlines the derivations of the PE computational model.
Validation of the PE method is given in Section III via several
representative examples.
II. PE MODEL
A. Scenario of the PE model
We work with A two-dimensional (2D) cylindrical
coordinates
, assuming the fields being independent of
the azimuth (
) direction, where
and
correspond to the
range and height, respectively. Time dependence is assumed as
. The simulation model shown in Fig. 1 composes an
analytical computation domain (I) and a PE domain (II). The
Parabolic Equation Method for Loran-C ASF
Prediction over Irregular Terrain
Dan-Dan Wang, Xiao-Li Xi, Member, IEEE, Yu-Rong Pu, Jang-Fan Liu, Li-Li Zhou