Printer Friendly

Analysis and comparison model for measuring tropospheric scintillation intensity for Ku-band frequency in Malaysia.


Radio-wave propagation through the Earth's atmosphere has a major impact on system design; several propagation effects increase in importance when comparing lower frequency bands, having a high degree of accuracy and comprehensiveness concerning their prediction (Agunlejika, et al., 2007). Propagation impairment regarding satellite communication links, especially in the Ku band and signal level fluctuation caused by attenuation due to rain and tropospheric scintillation, must be is carefully considered to ensure accurate link budgeting.

Tropospheric scintillation concerns rapid signal amplitude and phase fluctuation throughout a satellite link. It is caused by irregularities and turbulence in the first few kilometres above the ground, there by affecting atmospheric refractive index measurement (Mandeep et al., 2006). A link for propagation through the troposphere consists of combining random absorption and scattering from a continuum of signals along a path causing random amplitude and random scintillation in the waveform being received. Scintillation effect varies as time elapses and is dependent upon frequency, elevation angle and weather conditions, especially dense cloud. The greatest effect caused by tropospheric scintillation is signal fading, thereby acting as a limiting factor on system performance (Akhondi and Ghorbani, 2005).

This is why accurate prediction is important when evaluating a link budget, especiallyinhighlytroposphericscintillationconditions.Scintillation occurs continuously, regardless of whether the sky is clear or rainy. When it is raining, signal level fluctuation (known as scintillation) can change together with rain attenuation affecting signal level. Signal log-amplitude level will rise dramatically and such extreme level data should be carefully eliminated (Mandeep et al, 2006).

Data analysis

The measurement of data collected from a beacon satellite having 12 GHz frequency, 2.4m antenna diameter and 40[degrees] elevation angle were obtained by monitoring and collecting data from April 2008 to April 2009. Disanayake et al, (2002) have mentioned that most available beacon data has been analysed regarding clear sky conditions and this essentially removes the bulk of low-attenuation-producing phenomena. Table 1 gives measurement site specifications.

Signal attenuation due to rain is the most remarkable signal propagation effect in Ku-band frequency and this kind of loss due to the above can be greater than 15 dB over a short period of time (Otung, 1996). All data which has become changed due to attenuation caused byrain is eliminated.

Considering a clear sky (with or without rain), all data having a spike regarding extreme amplitude values due to rain attenuation has been removed by comparing it to rain gauge data values. Visual inspection was needed and performed for all data sequences to eliminate spurious and invalid data (Garcia, 2008). Full attention must be paid during inspection to ensure obtaining accurate result from studies. Scintillation variance values can be best described for scintillation intensity in the present study and have been calculated as the standard deviation of signal amplitude given in decibels (dB).

Comparison prediction model

Four prediction models were selected for this study: Karasawa (Karasawa et al, 2002), ITU-R (2009), Van de Kamp (Van de Kamp et al., 1999) Otung (Otung, 1996). The model so selected depended on its correlation with wet refractivity index value, and meteorological conditions, i.e. relative humidity (RH) and temperature, t ([degrees]C), these being suitable with scintillation data for a satellite beacon (Van de Kamp, 1998). Prediction model comparison was based on signal fading and enhancement. The chosen model was also able to predict long-term distribution propagation signals.

The Karasawa model

Karasawa has presented a prediction model for signal standard deviation regarding scintillation intensity as follows;

[[sigma].sub.pre] = [[sigma].sub.n] x [f.sup.0.45] x [square root of G(D.sub.a)]/sin [[theta].sup.1.3] for [theta] [greater than or equal to] 5 (1)

where [[sigma].sub.n] is normalised intensity, f is frequency in GHz, 9 is elevation angle and G([D.sub.a]) is antenna aperture averaging factor as given by:


where [lambda] is wavelength in m, is effective antenna diameter and L is the distance of the turbulent part of the path and can be determined as follows:

L = 2 b/[square root of [sin.sup.2] [theta] + 2 b/[a.sup.e] + sin [theta]] (3)

Concerning equation (1), Karasawa obtained the following expression for scintillation enhancement:


Signal fading can be expressed as:


The ITU-R model

The long-term tropospheric scintillation prediction model proposed by the International Telecommunication Union-Radio communication sector (ITU-R) was used for calculating the standard deviation of signal fluctuation due to scintillation. This model uses the wet term of earth refractivity wet N, regarding relative humidity and temperature, averaged at least once a month as input (Agunlejika et al., 2007). This model is applicable for frequencies ranging from 7GHz to 20 GHz and 4 to 32 elevation angles. The following equation can be used for the ITU-R prediction model;

[sigma] = [[sigma].sub.ref][f.sup.7/12] [g (x)[(sin [theta]).sup.1.2]] dB (6)


[sigma] = standard deviation (dB)

[[sigma].sub.ref] = reference standard deviation (dB)

g(x) = antenna averaging factor


[[sigma].sub.ref] = 3.6x [10.sup.-3] + [10.sup.-4] x [N.sub.wet] (dB) (7)

[N.sub.wet] = 3.732 x [10.sup.5] e/[T.sup.2] (8)

Referring to equation 6, scintillation fading can be calculated from the following equation for 0.01 [less than or equal to] p [less than or equal to] 50. No prediction model has been recommended by the ITU-R for scintillation enhancement.


The Van de Kamp model

The Van de Kamp prediction model represents a slight modification from the ITU-R model. Scintillation standard deviation for long-term distribution can be estimated from the equation given below;

[[sigma].sub.x] = [[sigma].sub.n][f.sup.0.45]g(x)/[(sin [theta]).sup.1.3] (10)

The percentage of time for scintillation intensity can be identified from the above equation, as in equations 11 and 12

[a.sub.1] (p) = -0.0515[(log p).sup.3] + 0.206[(log p).sup.2] - 181 log p + 2.81 (11)

[a.sub.2] (p) = -0.172[(log p).sup.2] - 0.454 log p + 0.274 (12)

Signal fading and enhancement can be determined as follows:

[A.sub.p] = [a.sub.1](p)[[sigma].sub.x] + [a.sub.2](p)[[sigma].sup.2.sub.x] (13)

[E.sub.p] = [a.sub.1](p)[[sigma].sub.x] + [a.sub.2](p)[[sigma].sup.2.sub.x] (14)

The Otung model

This model is similar to the ITU-R model, except for elevation angle dependent value which is sin [[theta].sup.-11/12] and this is shown as equation 15;

[[sigma].sub.x] = [[sigma].sub.xref x f 7/12 x G(D)]/sin[([theta]).sup.-11/12] (15)

Hence, fading and enhancement for signal level can be determined by using this equation:

A(p) = 3.6[[sigma].sub.x] exp(-0.00095/p-[0.4 + 0.002p] x ln(p)) (16)

E(p)= 3.17[[sigma].sub.x] exp(-0.00095p-[0.272-0.004p] x ln(p)) (17)

The analysis and comparison model

Figure 1 shows monthly cumulative distribution for scintillation variance considering average standard deviation of scintillation intensity over a one-month time period. Such variance was determined by considering clear sky conditions without rain. Percentage time value was lower than scintillation variance value for April 2008 and that for April 2009 was slightly higher than for the other month.

Figure 2 shows that average monthly scintillation distribution followed gamma distribution for long-term distribution data collection.



Long-term distribution data should be analysed more than once a month while only a few minutes are needed for short-term data analysis. Figure 2 gives values regarding negative state for signal level enhancement while correct or positive state is for signal level fading. It obviously shows that variation in variance scintillation value for fading and enhancement was not equally likely. Signal fading had a long tail compared to enhancement and the shape was not symmetrical, as has been mentioned by Van de Kamp (1998).

Fading and enhancement represent two types of scintillation signal level. Bothhavetheirownuseandfunctionalitieswhichcanhavealargeeffectonthe propagation of a signal being transmitted through the atmosphere. When propagation signals are affected by rain, especially during the raining season, fading value will suffer a drastic change due to changes in signal amplitude. However, the enhancement value is not affected by rain or can become negligible.


Figure 3 represents cumulative distribution for signal level fading and enhancement.

Variance distribution for fading was slightly higher when comparing enhancement value for the lower percentage of time. Such cumulative distribution was for a local data study with specific meteorological conditions due to geographical conditions. 26[degrees]C temperature and 76% humidity were used in the present study.


Prediction model selection was based on their relationship to meteorological conditions. Comparing these four models showed that the Karasawa model was the best model for predicting scintillation data intensity, as shown in Figure 4 for scintillation signal fading (~26[degrees]C temperature (t) and 76% relative humidity, (RH)).

Figure 4 shows that the Karasawa model gave good prediction, having 0.007dB minimum signal variance, 1.8% of this referring to the measured data. The Karasawa was thus a suitable model for predicting local data regarding scintillation intensity for signal fading compared to the other models while the Otung model did not perform well in predicting scintillation data (0.12dB and 35% from measured data as reference).

However, only three models performed well regarding signal enhancement, as shown in Figure 5. This was because no prediction model has been proposed by the ITU-R for signal enhancement (ITU-R 2007); only the Karasawa, Van de Kamp and Otung models will thus be compared. Figure 5 shows signal enhancement, at ~26[degrees] C and 76% humidity value.

This comparison obviously showed that the Karasawa model also performed well for predicting signal level enhancement regarding scintillation data intensity. A small difference regarding variance value with 0.0052dB and 2.6% as reference compared to the other models. The Otung model was the worst model (0.0414dB and 20.96% reference values).



Tropospheric scintillation predication models have been reviewed and evaluated, including models for predicting signal log-amplitude cumulative distribution and models for predicting scintillation intensity. This tropospheric scintillation intensity study responded to the requirement for better understanding of propagation impairment in satellite communication systems. Better understanding can produce better system design. This study thus concluded that the Karasawa prediction model can be best used for predicting overall propagation impairment regarding scintillation on the Malaysian propagation path.


The author would like to acknowledge the Universiti Kebangsaan Malaysia, Universiti Sains Malaysia, MOSTI grant Science Fund (01-01-92-SF0670), UKM-GGPM-ICT-108-2010, the Association of Radio Industry Business (ARIB) of Japan for providing the instruments used for collecting the data and the Research University Postgraduate Research Grant Scheme (USM-RU-PGRS).


Agunlejika, O., T.I. Raji and O.A. Adeleke (2007). "Tropospheric scintillation prediction for some selected cities in Nigeria's tropical climate", International Journal of Basic and Applied Science, 9, 10, 12-19.

Akhondi, M. and A. Ghorban (2005). "Long-Term cumulative distribution modeling of tropospheric scintillation for the earth-satellite links in the 40/50 GHz band", 4th WSEAS International Conference on Electrical, Hardware., Wirel. and Optical Commmunication., 1, Wisconsin, USA, 56-59.

Dissanayake A., J. Allnutt and F. Haidara (2002). "A prediction model that combines rain attenuation and other propagation impairments along earth-satellite paths", IEEE Transaction on Antenna and Propagation, 45, 10, 1546-1558.

ITU-R Recommendation P.618-9 (2009). "Propagation data and prediction methods required for the design of earth-space telecommunication systems",

Karasawa, Y., M. Yamada and J. Allnutt (2002) "A new prediction method for tropospheric scintillation on earth-space paths" IEEE Transaction on Antenna and Propagation, 36, 11, 1608-1614.

Mandeep, S. J. S., I.S.H Syed,, I. Kiyoshi, T. Kenji and I. Mitsuyoshi (2006). "Analysis of troposphericscintillationintensityonearthtospaceinMalaysia",AmericanJournal of Applied Science, 3, 9, 2029-2032.

Otung, I. E. (1996). "Prediction of tropospheric amplitude scintillation on a satellite link". IEEE Transaction on Antenna and Propagation 44, 9, pp 1600-1608.

Van de Kamp, M.M.J.L, J.K Tervonen, E.T. Salonen, and J.P.V.P Baptista,. 'Improved modelsforlong-termpredictionoftroposphericscintillationonslantpaths'.IEEE Transaction on Antenna and Propagation, 47, 12, 249-260.

Van de Kamp, M.M.J.L. (1998) "Asymmetric signal level distribution due to tropospheric scintillation", Electronic Letters, 34, 17, 1145-1146.

Manuscript received: 25/01/2011

Accepted for publication: 28/05/2011

JS Mandeep, RM Zali

Department of Electrical, Electronic & Systems Engineering, Faculty of Engineering & Built Environment, Universiti Kebangsaan Malaysia, 43600, UKM Bangi, Selangor Darul Ehsan, Malaysia Email:,
Table 1. Satellite specifications

Ground station location 5.170N, 100.40E
Beacon frequency 12.255 GHz
Elevation angle 40.10
Polarisation Horizontal
Antenna configuration Offset parabolic
Antenna diameter 2.4m
Satellite position 1440E
Antenna height 57m above sea level
COPYRIGHT 2011 Universidad Nacional de Colombia, Departamento de Geociencias
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:ATMOSPHERE
Author:Mandeep, J.S.; Zali, R.M.
Publication:Earth Sciences Research Journal
Date:Jul 1, 2011
Previous Article:Tunnelling in urban areas by EPB machines: technical evaluation of the system.
Next Article:Scanning electron microscope description of the Afowo oil sand deposits in south-western Nigeria.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters