See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/266522655

Atmospheric influences on satellite

communications

Article in Przeglad Elektrotechniczny · January 2011

CITATIONS READS

4 182

4 authors, including:

Zaffar Haider Janjua Jamal Nasir

University of Milan Faculté des Sciences et Techniques Fès
7 PUBLICATIONS 63 CITATIONS 29 PUBLICATIONS 41 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Multi-beam phased arrays for 5G handheld devices View project

Phased array antenna View project

All content following this page was uploaded by Jamal Nasir on 23 April 2016.

The user has requested enhancement of the downloaded file.

 Muhammad ZUBAIR1, Zaffar HAIDER2, Shahid A. KHAN3, Jamal NASIR4
COMSATS Institute of Information Technology, Islamabad (1,2,3), COMSATS Institute of Information Technology, Abbott Abad (4)

Atmospheric influences on satellite communications

Abstract. Among other atmospheric regions, ionosphere, which is ionized region of the atmosphere, is considered to impose serious limitations on
satellite communication. At higher frequencies, radio waves pass through the ionosphere and are attenuated due to the free electrons present in
ionosphere. This paper discusses atmospheric effects on high frequency radio waves illustrating the attenuation and losses it may come across like
attenuation due to atmospheric gases, rain, clouds, beam spreading loss and noise temperature. Data used to summarize the influence of
atmospheric phenomena is obtained from ITU-R models and processed using MATLAB.

Streszczenie. Spośród różnych warstw atmosfery szczególnie jonosfera wpływa niekorzystnie na komunikację za pośrednictwem satelitów. Przy
wyższych częstotliwościach fale radiowe są silnie tłumione w jonosferze. Artykuł przedstawia analizę wpływu czynników atmosferycznych w
jonosferze na tłumienie i straty w radiowej komunikacji wysokoczęstotliwościowej. (Wpływ atmosfery na jakość komunikacji satelitarnej)

Keywords: Radio propagation, Ionosphere, Rain attenuation, cross polar discrimination.

Słowa kluczowe: propagacja fal radiowych, komunikacja satelitarna

Introduction In most countries, water vapour density is not available

The ionosphere is the upper part of the atmosphere for propagation interest; hence it can be calculated from
where sufficient ionization exists to influence radio wave concurrent measurements of temperature and relative
propagation. The ionosphere usually consists of two layers: humidity using:
the E layer which is about 80 to 113 km above the earth’s
surface and reflects radio waves of lower frequency. Above (1) 
  (U / 5.752)  6 .10 10 9.834  ,
E layer is F layer which reflects higher frequency radio where U is the relative humidity in percent. θ is the inverse
waves. The F layer is then further sub divided into F1 and temperature constant, given by:
F2 layers. The F1 layer is lower portion of F layer and exists
from 150 to 200 km above the earth’s surface, whereas F2   300 / T0
layer is the upper portion and exists at a height of 200 to T0 is the temperature at the surface in Kelvin [8].
500 km. F2 layer is mainly responsible for reflection of HF The total gaseous attenuation in the atmosphere A over
waves during day and night. Since the ionization is mainly a path length r0 (km) is given by:
caused by solar radiations, it is dependent on location, time 0
A    0 (r )   w (r )dr , dB
[1].
of the day, season and sunspots (2)
Radio waves propagating through ionosphere 0
experience different attenuation mechanisms such as where β0 and βw are the attenuation coefficients (in decibels
absorption, reflection, refraction, scattering, polarization, per kilometer) for oxygen and water vapour, respectively [9].
group delay and fading/scintillation. In the region other than
ionosphere i.e. troposphere, stratosphere etc. radio waves
loses its energy mainly due to absorption, cloud and rain
attenuation, attenuation due to snow, hail and fog. Rain is
considered to be the major cause of attenuation at
frequencies above 10 GHz.
As reported in literature atmosphere contains free
electrons, ions, and molecules and their interaction with
radio waves depend strongly on frequency, so as the frequ-
ency increases, the effect of attenuation also increases.
Rest of the paper is organized as follows: Section II
discusses attenuation due to atmospheric gases. Section III
discusses attenuation due to precipitation and clouds.
Section IV describes beam spreading loss. Section V
describes cross polarization at different rain rates, elevation
angles and tilt angles. Finally, Section VI gives the
conclusion. Fig 1: Specific Attenuation due to atmospheric gases for a standard
atmosphere in UHF, SHF and EHF bands.
Attenuation due to Atmospheric Gases
Attenuation by atmospheric gases at microwave and As shown in Fig-1, dry air has an oxygen absorption line
millimetric frequencies is mainly due to oxygen and water at 60 GHz. The first absorption band at 22.2 GHz is due to
vapor absorption. Oxygen possesses a permanent water vapour, followed by absorption at 60 GHz due to dry
magnetic moment and because of the interaction of this air and at 118 and 123 again due to water vapour. The
moment with the magnetic field of the wave, absorption of ‘Atmospheric windows’ between these absorption bands are
wave energy takes place [1]. available for practical earth-space communications.
At frequencies below 3 GHz, path attenuation due to It is evident from Fig-1 that below 22.3 GHz, the specific
atmospheric gases, rain and clouds is small and is often attenuation increases with frequency tremendously and it
neglected. Whereas oxygen and water vapour in the lower can be more than 10 times higher at 15 GHz than at 2 GHz.
atmosphere significantly affect path attenuation at higher Also, the gaseous absorption is less than 1 dB for most
frequencies. As the effect is highly frequency dependent so paths below 100 GHz as indicated in Fig.1.
the attenuation due to atmospheric absorption in some Taking into account the relative contribution, it is
frequency bands is much greater than in others. obvious that there will be more attenuation in presence of

PRZEGLĄD ELEKTROTECHNICZNY (Electrical Review), ISSN 0033-2097, R. 87 NR 5/2011 261

water vapour than in dry air because of the presence of Rain attenuation is a key limiting factor in using high
more molecules in water vapour. Water vapour is a polar frequency bands in satellite and terrestrial microwave
molecule with an electric dipole resulting in two absorption systems [1]. Rain drops both absorb and scatter radio-wave
lines in the microwave region at 22.2 GHz and 183.3 GHz, energy. Very intense rain rate may cause link outage. If the
where as Oxygen molecule has a permanent magnetic rain drop size approaches half the wavelength of the signal
moment that produces multiple absorption lines that spread in diameter, the signal will be attenuated. Higher
out between 50 and 70 GHz. frequencies exhibit more attenuation than lower frequencies
In conclusion, Atmospheric oxygen and particularly due to smaller wavelength as shown in Fig. 2.
atmospheric water vapour cause a minor level of Clouds are also an important source of attenuation at
attenuation to satellite signals. The effects generally higher frequencies. Due to diverse nature of clouds,
increase with frequency and are greatest near lines in the attenuation of different intensities may occur. Each type of
absorption spectra for each molecule. The line of most cloud has different water droplet concentration. Clouds
interest normally is the water absorption line at 22.3 GHz as having ice crystals cause less attenuation. Non precipitating
we normally use the SHF band i.e. 3 to 30 GHz. Besides clouds are also not very significant as the liquid content is
frequency the amount of absorption depend on the humidity too low to cause much absorption of energy, and the
(water vapour concentration), the elevation angle, pressure droplets are too small to scatter the energy and also they
and temperature. are spherical so they can not cause cross polarisation. In
warmer climates the clouds are thicker so cloud attenuation
Attenuation due to Precipitation and Clouds may be higher.
The strength of satellite signal may be degraded or Rain degrades the performance of a satellite
reduced under rain conditions; in particular radio waves communication system by increasing the noise temperature
above 10 GHz are subject to attenuation by molecular of the earth station antenna.
[7]
While raining it receives
absorption and rain [5]. Presence of rain drops can severely thermal radiation from rain drops which cause an increase
degrade the reliability and performance of communication in the overall noise temperature.
links. Attenuation due to rain effect is a function of various
parameters including elevation angle, carrier frequency,
height of earth station, latitude of earth station and rain fall
rate. The primary parameters, however, are drop-size
distribution and the number of drops that are present in the
volume shared by the wave with the rain. It is important to
note that, attenuation is determined not by how much rain
has fallen but the rate at which it is falling [10].
The propagation loss due to rain is given by:
Pr (0)
(3) L  10 log ,
Pr (r )
where P0 is the signal power before the rain region, Pr is the
signal power after the rain region, and r is the path length
through the rain region. Fig 3: Sky noise temperature as seen by the antenna.
The propagation loss due to rain attenuation is usually
The antenna collect noise from ground, atmosphere
expressed by specific attenuation  , in decibels per
(whether cloud or rain), and extraterrestrial sources.
kilometre, so propagation loss is: Antenna noise temperature varies with elevation angle
antenna size, frequency and weather conditions.
(4) L   lr
where  is specific attenuation in dB/km and l r is rain path Beam Spreading Loss
Due to the regular decrease of the radio refractive index
length in km.[11] of the atmosphere with height, downward ray bending is
[12]
Based on ITU-R specific attenuation model it is found produced, which increases by reducing the elevation angle
that γ depends only on rainfall rate, measured in millimetres of the ray. The effect is significant at elevations below about
per hour. From this model, the usual form of expressing γ 3° [4].
is:
(5)   aR b [ dB / km ]
where a and b are frequency dependent coefficients.

Fig 4: Loss with respect to elevation angle as a function of latitude.

Rays at the top and bottom of the antenna main beam

Fig 2: Rain attenuation as determined by ITU-R model and Simple travel with slight different elevation angles and an additional
Attenuation Model (SAM) at frequencies above 10 GHz. divergence of the beam is produced in the vertical plane

262 PRZEGLĄD ELEKTROTECHNICZNY (Electrical Review), ISSN 0033-2097, R. 87 NR 5/2011

due to the resulting differential ray bending. There is no
increase in the divergence of the beam in horizontal plane.
In satellite communications, beam-spreading loss results
from the spreading of the earth-satellite signals as they
pass through the earth’s atmosphere. Beam spreading loss
is maximum at lowest elevation angle and decrements by
increasing the elevation angle as shown in Fig. 4. At 3°, the
loss becomes constant. It is also clear from the figure below
that the loss is comparatively smaller if the antenna is
closer to the equatorial region than the one which is far from
equator.

Cross Polarization Discrimination

Large raindrops in the rain are not spherical but flatten
and fall with their major axis almost horizontal. The
horizontal component of the wave is thus more attenuated
when it propagates through the rain. If we recombine the
horizontal and vertical components at any point to
reconstruct the wave, we will see that its polarization has
rotated towards the component. Thus, a cross polarization
[3]
component has come to existence . Fig.6 Vector relationship for a depolarizing medium. a) co- and
Depolarization is induced by two factors: a) Rain and b) cross-polarized waves for linear transmission and b) classical
multipath propagation. Multipath induced depolarization is model for a canted oblate spherical rain drop.
[13]

generally limited to terrestrial links. The major

depolarization on satellite paths is caused by rain and ice. The transmission characteristics of the oblate spheroids
The wave while passing through the anisotropic medium is specified by transmission coefficients in the I and II
exhibits attenuation and phase shift and thus its polarization directions as:
state is altered, such that power is transferred from desired (8)  I  e  ( AI i1 ) L
polarization state to the undesired orthogonal polarization
state, resulting in interference.  II  e  ( AII i II ) L
where AI and AII are attenuation coefficients, and ФI and
ФII are phase coefficients in I and II directions respectively.
The received waves are then:
(9) E RI   I ETI
E RII   II ETII

The resulting received waves in the directions 1 and 2

can be obtained by combining (7), (8) and (9):
(10) E R1  11 E1   21 E 2
E R 2  12 E 2   22 E 2
Where β11, β21, β12 and β22 are polarization coefficients
defined by
Fig 5: Variation of XPD with frequency and percentage of an
average year.
(11)  11   I cos 2    II sin 2 
Cross polarization Discrimination (XPD) depends upon  22   I sin 2    II cos 2 
attenuation which mainly depends upon the amount or the
  II   1  2
volume of water present in the atmosphere [3].  12   21    sin 
 2 
(6) XPD = U – V log10 (A) dB
The XPD’s for vertical and horizontal transmission will
Where: U = 30 log(f) – 40 log(cos ε) – 10 log ½ [1 – cos (4τ) then be
–k 2
e m and and: ε - is the satellite elevation angle, τ - is the 
local polarisation tilt angle, km2 relates to the variance 1  II tan 2 
–k 2 2 11 1
of the canting angle distribution ( m = 0.0024 θm ). (12) XPDv  20 log  20 log
As it can be seen from Fig. 5 that Cross Polarization  12   II 
Discrimination increases with the increase in frequency and   1 tan 
decreases with the percentage time of an average year.  1 
Consider two orthogonally polarized linear waves are  II
passing through a volume of canted oblate spheroids as  tan 2 
 22 1
shown in Fig-6. The drops are canted at an angle φ at the (13) XPD H  20 log = 20 log
horizontal in the directions I and II, with minor and major  21   II 
axes as a and b, respectively. Equations (7-13) determine   1 tan 

 1 
XPD for both vertical and horizontal polarization transmitted
[13]
waves . The transmitted waves are: Both 11 and 22 are independent of , and since 12 = 21,
the cross polarized components resulting from positive and
(7) ET 1  E1 cos   E1 sin  negative canting angles will cancel out each other.
ETII  E1 sin   E 2 cos 

PRZEGLĄD ELEKTROTECHNICZNY (Electrical Review), ISSN 0033-2097, R. 87 NR 5/2011 263

 The XPD for circular transmitted polarization (right hand overcome these problems a model is needed to be
or left hand circular polarization) is expressed in terms of designed to use 1-hr data without converting to 1-min rain
polarization coefficients as [13], rate data.
     II   I i 2  The results of both models are sometimes not consistent
(14) XPDc  20 log 11 e i 2   20 log e  due to the reason that these models are originally designed
  12   45o    II   I 
  for Europe and America, so we can rely only on these
models where real data is not available. For more accurate
where e i 2 is the mean of e i 2 taken over the canting results, data from local meteorological departments can be
angle distribution. obtained and based on the empirical data local model
As, higher the value of XPD, less will be the should be designed which will be more consistent with real
depolarization factor and better will be the received signal. time statistics in the region.
So increase in frequency increases the received signal
quality in this case. REFERENCES
nd
[1] Martin Hall, Les Barclay, Propagation of Radiowaves, 2
edition, The Institute of Electrical Engineers, London, UK.
[2] ITU-R Recommendation P.676-5, Attenuation by atmospheric
gases, 2001.
[3] ITU-R Recommendation P.618-7, Propagation data and
prediction methods required for the design of Earth-space
telecommunication systems, 2001.
[4] ITU-R Recommendation P.834-4, Effects of tropospheric
refraction on radiowave propagation, 2003.
[5] ITU-R Recommendation P.837-5, Characteristic of
precipitation for propagation modeling, 2007.
[6] Dong Y.Choi, Rain attenuation prediction model by using the 1-
hour rain rate without 1-minute rain rate conversion, IJCSNS
Iinternational Journal of Comp. Science and Network Security,
March 2006, Vol. 6 No. 3A.
[7] ITU-R Recommendation, P.618-8, Propagation data and
prediction methods required for the design of Earth-space
Fig 7: XPD varies with path elevation angle and the rain attenuation
telecommunication systems, 2003.
exceeding the required percentage of time (Ap)
[8] Glen Feldhake, Estimating the attenuation due to Combined
Increasing tilt angle decreases the cross polarization Atmospheric Effects on Modern Earth-Space Paths, IEEE
component as long as it is below 45° as shown in Fig. 8. Antennas and Propagation Magazine, August 1997, Vol. 39,
No. 4.
Since Cross polarization discrimination is a function of
[9] CCIR, Report 719, Attenuation by atmospheric gases,
copolar attenuation, it degrades by increasing attenuation. Recommendations and reports to the CCIR, 1978, Vol. 5, pp.
Also it degrades at a given attenuation with increasing 97-102.
frequency. XPD for vertical polarization shows significantly [10] Hany E. Green, Propagation Impairment on Ka-band
good results than for horizontal polarization. Similarly, XPD SATCOM links in Tropical and Equatorial Regions, IEEE
is better for linear polarization than circular polarization [3]. Antennas and Propagation Magazine, April 2004, Vol. 46, No.2.
[11] Shkelzen C., Kresimir M., Rain Attenuation at Low Earth
th
Orbiting Satellite Ground Station, Presented in 48
International symposium ELMAR-2006, 07-09 June 2006,
Zadar, Croatia.
[12] ITU-R Recommendation P. 838-1, Specific Atteuation model
for Rain for use in Prediction Methods, Geneva, 1992.
[13] Louis J. Appolito, Radio Propagation for Space Communication
Systems, In Proceedings of the IEEE, June 1981, Vol. 69, No.
6.
[14] T. S. Chu, Rain Induced Cross-polarization at centimeter and
millimeter wavelengths, Bell. Sys. Tech. J., Oct. 1974, Vol. 53,
no.8, pp. 1557-1579.

Authors
Muhammad Zubair, Department of Electrical Engineering,
COMSATS Institute of Information Technology, Park Road,
Islamabad. Email: muhdzubair@comsats.edu.pk
Zaffar Haider, Department of Electrical Engineering, COMSATS
Institute of Information Technology, Park Road, Islamabad. Email:
zaffar_haider@comsats.edu.pk
Fig 8: XPD vs. tilt angle of linearly polarized electric Field.
Prof. Dr. Shahid A. Khan, Department of Electrical Engineering,
Conclusion COMSATS Institute of Information Technology, Park Road,
This paper describes how radio propagation is affected Islamabad. Email: shahidk@comsats.edu.pk
Jamal Nasir, Department of Electrical Engineering, COMSATS
by various factors in stratosphere and ionosphere. Both Institute of Information Technology, Abbottabad. Email:
ITU-R and SAM models were used to determine the jamalnasir@ciit.net.pk
influence of atmosphere on radio waves in different weather
conditions.
The ITU-R model suggests that a 1-min interval must be
used to measure rain fall in order to determine rain fall rate,
which is very difficult in practice if we want to maintain 1-
min data for several years. Furthermore, the 1-hr rain rate
data need to be converted to 1-min rain rate data so that it
can be applied to the existing rain attenuation models. To

264 PRZEGLĄD ELEKTROTECHNICZNY (Electrical Review), ISSN 0033-2097, R. 87 NR 5/2011

View publication stats