Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Predicting atmospheric propagation of radio frequency waves using meteorological data and the split-step parabolic equation

Nilsson, Erik LU (2026) EITM01 20252
Department of Electrical and Information Technology
Abstract
This thesis, along with the Propagation Predictor written in MATLAB, offers a
quick and easy way of predicting how Electromagnetic (EM)-waves will propagate
given the current atmospheric situation. While the thesis is written with the goal
of explaining choice and phenomena, the Propagation Predictor is built with the
goal of not requiring deep knowledge of neither EM-waves nor meteorology.
Meteorological data is used to predict tropospheric propagation of EM-waves
in the Radio Frequency (RF) spectrum. Theory for EM-waves, meteorology and
atmospheric refractivity is presented, as well as the mathematics required for the
simulation.
The simulation is done using a wide angle approximation Split Step Parabolic
Equation (SSPE).... (More)
This thesis, along with the Propagation Predictor written in MATLAB, offers a
quick and easy way of predicting how Electromagnetic (EM)-waves will propagate
given the current atmospheric situation. While the thesis is written with the goal
of explaining choice and phenomena, the Propagation Predictor is built with the
goal of not requiring deep knowledge of neither EM-waves nor meteorology.
Meteorological data is used to predict tropospheric propagation of EM-waves
in the Radio Frequency (RF) spectrum. Theory for EM-waves, meteorology and
atmospheric refractivity is presented, as well as the mathematics required for the
simulation.
The simulation is done using a wide angle approximation Split Step Parabolic
Equation (SSPE). Polarization is ignored and the treatment of reflections is simplified.
Attenuation from air, precipitation, and condensation is calculated from
the same data as the propagation.
The MATLAB program written (the Propagation Predictor) demonstrates a
feasible approach for how meteorological data can be used in predicting the propagation
of RF based sensor systems. While the SSPE is an established approach,
previous works suggest that the vertical fidelity of meteorological data is too low
for accurately recreating the atmospheric layers. In addition to this, reflections,
especially for rough terrain, risk being to simplified for accuracy. Comparison to
a similar model shows deviations, indicating that the simulation is not entirely
accurate. (Less)
Popular Abstract
If you go to the ocean and look out along the horizon, you might see boats. They
can appear behind, on top of or in front of the horizon, and on some days they
might even appear to be floating above it. This illusion, or mirage, comes from
light being bent, meaning that the boat you see is not in a straight line from
your eyes. These mirages are standard for radars, but sometimes the mirage has
a larger effect than expected. This is called an anomaly.
For you, seeing a flying boat over the horizon doesn’t matter much. You know
that boats can’t fly, and can easily recognize it as a mirage. For radars this is
harder, and it might not even be a boat that’s observed; seeing an aeroplane, how
can one be certain of its altitude?
In... (More)
If you go to the ocean and look out along the horizon, you might see boats. They
can appear behind, on top of or in front of the horizon, and on some days they
might even appear to be floating above it. This illusion, or mirage, comes from
light being bent, meaning that the boat you see is not in a straight line from
your eyes. These mirages are standard for radars, but sometimes the mirage has
a larger effect than expected. This is called an anomaly.
For you, seeing a flying boat over the horizon doesn’t matter much. You know
that boats can’t fly, and can easily recognize it as a mirage. For radars this is
harder, and it might not even be a boat that’s observed; seeing an aeroplane, how
can one be certain of its altitude?
In order to better understand the mirages, we can use weather data for the
place we’re looking at. The weather is what causes the anomalies, so by simulating
how light will travel through the air we get a clearer view of what is real.
SMHI has a lot of data available on their website, and the type called MEPS contains all that is needed and more. By choosing to extract temperature, pressure,
humidity, cloud and rain, we can calculate how much the air should bend the light,
given the current weather. Here we can also modify the values a bit, in order to
treat the earth as flat. However, guessing how the light is bent just by looking at
different values is often just that - guessing - so in order to get better predictions we need to simulate the light.
How light behaves is well known, but using the fully correct models is very
difficult. Instead, we set limitations to make calculations easier. First, we say that light only travels in a cone; instead of a lightbulb we simulate a headlight. We then say that this headlight can only point in one direction (forward) and that it can’t be tilted up or down too much. Finally, we put a piece of paper along the
light beam from the headlight, and choose to only simulate how it looks on that
piece of paper. The end product of this is slices showing how strong the light is at different distances and different heights. Going back to the initial analogy, this can give an indication of where the flying boat is most likely to be. (Less)
Please use this url to cite or link to this publication:
author
Nilsson, Erik LU
supervisor
organization
course
EITM01 20252
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Radio frequency, Atmosphere, Propagation, Sensor, Radar, Parabolic equation, SSPE, SMHI, Weather
report number
LU/LTH-EIT 2026-1119
language
English
id
9225967
date added to LUP
2026-05-20 11:01:27
date last changed
2026-05-20 11:01:27
@misc{9225967,
  abstract     = {{This thesis, along with the Propagation Predictor written in MATLAB, offers a
quick and easy way of predicting how Electromagnetic (EM)-waves will propagate
given the current atmospheric situation. While the thesis is written with the goal
of explaining choice and phenomena, the Propagation Predictor is built with the
goal of not requiring deep knowledge of neither EM-waves nor meteorology.
Meteorological data is used to predict tropospheric propagation of EM-waves
in the Radio Frequency (RF) spectrum. Theory for EM-waves, meteorology and
atmospheric refractivity is presented, as well as the mathematics required for the
simulation.
The simulation is done using a wide angle approximation Split Step Parabolic
Equation (SSPE). Polarization is ignored and the treatment of reflections is simplified.
Attenuation from air, precipitation, and condensation is calculated from
the same data as the propagation.
The MATLAB program written (the Propagation Predictor) demonstrates a
feasible approach for how meteorological data can be used in predicting the propagation
of RF based sensor systems. While the SSPE is an established approach,
previous works suggest that the vertical fidelity of meteorological data is too low
for accurately recreating the atmospheric layers. In addition to this, reflections,
especially for rough terrain, risk being to simplified for accuracy. Comparison to
a similar model shows deviations, indicating that the simulation is not entirely
accurate.}},
  author       = {{Nilsson, Erik}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Predicting atmospheric propagation of radio frequency waves using meteorological data and the split-step parabolic equation}},
  year         = {{2026}},
}