Estimation of Rain Attenuation at EHF bands for Earth-to-Satellite Links in Bangladesh

Estimation of Rain Attenuation at EHF bands for Earth-to-Satellite Links in Bangladesh Md. Sakir Hossain Dept. of Electroni c and Telec ommunicati ons Engineering International Islamic University Chittag ong Chittagong, Bang ladesh Email: shakir.rajbd@yahoo.com Md. Atiqul Islam Dept. of Electri cal and Electronic Engineering International Islamic University Chittag ong Chittagong, Bang ladesh Email: atiq_atrai@yahoo.co m Abstract —Due to heavy c ongestion in low er frequency bands, engineers are looking for new frequency bands to support new services that require higher data rate s, which in turn needs broader bandwidths. To meet this requirement, extremely high frequency (EHF), particularly Q (36 to 46 GHz) and V (46 to 56 GHz) bands, is the best viable solution because of its complete availability. The most serious c hallenge the E HF band poses is the attenuation caused by rain. This paper investigates the effe ct of the rain o n Q and V bands’ performances in Banglade shi climatic conditions. The rain atte nuations of the two bands are predicte d for the four main regions of Bangladesh using ITU rain attenuation model. The measured rain statistics is used for thi s prediction. It is observed that the attenuation due to rain in the Q/V band reaches up to 150 dB which is much highe r than that of the currently used K a ba n d. T he v ar i ab i li t y of th e r ai n at t en u at i on is also investig ated over diffe rent sessions of Bangladesh. The attenuation varies from 40 dB to 170 dB depending on the months . F in a l l y, t h e a m o un t o f ra i n fa d e r eq u i re d t o c o m p en s a te t he h i g h rain attenuation is also predicted for d ifferent elevat ion angl es. Keywords—Satellite; Bangladesh; Attenuation; Rain; EHF bands I. I NTRODUCTION The usage of w ireless comm u nication is following a very sharp uptrend. Particularl y, the extensive use of multim edia communicat ion makes frequency bands s carce. This m akes lower freque ncy bands very congested, an d the need for attem pt to use higher frequency bands is imperative. To materialize the global gigab yte wireless co nnectivity, satellite communication is deemed as one of the most important too l. In satellite communicat ion, the most widely used frequency band is C band, which is al ready congested heavil y. The Ku and Ka bands will become saturated very soon becau se the use of these bands is increasing rapidly . This motivates engi neers to look for higher frequency bands f or getting av ailable bandwidth f or future satellite communicatio n to cope with the upcoming tremendous demand of new multimedia services. The two bands of EHF band, namely Q and V bands, are being seen as the possible solutio n to this tremendous scarcity of bandwid th. EHF bands have a num ber of advantages, such as broader bandwidth, low interference due to high directivity, low transmission po wer due to high gain antenna, small antennas and so on [1]. Few examples of EHF satellite systems are INTALSAT F1 and F2, Sicral, Alphasat [1]. However, frequency bands above 10 GHz do not appear as solely blessing, rather these are accompanied with severe rain attenuation. The rough estimate of rain attenuation increase is almost the square of the frequency a b ove C band [2]. T o design a satellite communication system in Q/V band s, the estimation of rain attenuation in these bands is a prior task for effici en t link design. The rainfall rate is not same all over the world, instead it varies si gnificantly. The amount of rai n attenuation depends on the rainfall rate. So the rain attenu ation is n ot f ixed across the world. The rain attenuations in different frequency bands are different. The effects of weather related issues on microwave link in th e Bang ladeshi cli matic con dition are investigated in [3]-[7]. The rain attenuation for Bangladeshi climate in C, Ku and Ka bands has been predicted in [3], [6], [7]. Considering the uptrend of the congestion in the low er frequency b ands, it is necessary to perform a feasibility study of using highe r frequency bands. In this paper, the rain attenuatio ns in the Q and V bands are predicted for Bangladeshi climate to facilitate satellite link design in Bangladesh . It is observ ed that the attenuatio n in the Q and V bands are 50 dB and 100 dB higher compared to that in Ka bands for 0.001% time. Th is requires a very large fade margin to keep the received signal in the EHF band acceptable. The required fa de margins are a round 150 dB and 110 dB respecti vely in the Q and V bands, respectively, for 0.01% time of the year. Sylhet r egion requires the highest fade margin, which is about 20 dB higher compared to the northern region Rajshahi, the least affected region. In addition, the attenuation increases four times in rainy season compared t o winter season. II. R AIN R ATE D ISTRIBUTIO N IN B ANGLADESH Bangladesh has subtropical mons oon climate characterized by wide seasonal variations in rainfall. Heavy rain fall is a featu re of Banglades h. There i s a significant variation in the annual rainfall across Bangladesh; this is highest in Sylhet due to it s location i n the so uth of the foothills of the Himalayas, while th e minimum is experienced in relatively dry western region of Rajshahi. Including the rain attenuatio n in the mentioned two regions, the attenuation for Dhaka and Chittagong, two most 589 International Conference on Electrical, Computer and Communication Engineering (ECCE), February 16-18, 2017, Cox’s Bazar, Bangladesh Fig. 1 Cumulative rain distribution of Bangladesh by R-H rain m odel industrial regions where most of the earth stations are located , are predicted here. The outage of a satellite system depends considerabl y on the signal att enuation due to the rain, whi ch i n turn depends on instantaneous rain rate. But the rain rate is n ot constant over the year. So cumulati ve distribution (CD) of the rain acro ss the year is used to predict rain attenuation. Here the annual rainfall statistics for forty years (from 1968 to 2008) of four differe nt regions are coll ected from Banglades h agricul tur e research council (BARC) [8], an d the average annual rainfall is computed from the data. Then, the CD of rain rate is found using Rice-Holm berg Rain Model (R-H r ain model) [9]. The parameters nee ded for R-H model include Thu nderstorm Rat io β (amount of convective rain compared to the total rainfall) an d the average annual rainfall depth M. The values of these parameters are shown i n Table I, where β is taken from the m ap in [9]. TABLE I. P ARAMETERS OF R-H R AIN M ODEL Parameters Dhaka Chittagong Rajshahi Sylhet M (mm) 2124 2887 1545 4101 β 0.5 0.5 0. 5 0.5 This model gives the rain rate CD at 1-minute integration ti me which is shown in Fig. 1 for the parameters mentioned i n Tabl e I. The 0.01% rain rates (R 0.01 ) of four regions taken fr om Fig. 1 a r e l i s t e d i n T a b l e I I , w h e r e t h e m a x i m u m R 0.01 , w h i c h i s 141.661 mm/h, reported in Sylhet, while the minim u m of 109.1496 mm/ h in Raj shahi. Howev er, the predicted 0.01% rain rate in Bangladesh is 95 mm/h according t o the Recommendation ITU-R P.837-6 [10 ]. A significant difference is observed between the measured R 0.01 and ITU predicted R 0.01 . TABLE II. R 0.01 I N D IFFERENT R EGIONS Region R 0.01 (mm/h) Dhaka 119.7673 Chittagong 129.9933 Rajshahi 109.1496 Sylhet 141.6991 ITU Map 95 Fig. 2 Schematic p resentation of an Earth-space path gi ving the para meters to be input into the attenuation predictio n process [17]. III. R AIN A TTENUAT ION P REDICTION M ODEL There are a number of r ain attenuation predic tion models [11]. Of t hem ITU-R [12] and CCIR [13] are designed for world- wide application for wide range of frequencies, rain climates, and elevation angles. The rest of the models mentioned in [11] are designed to meet the local needs. However, ITU-R rain attenuation prediction model is m ore widely used than CCIR because of its more accu racy. In this paper , ITU-R m odel will be used for predicting the attenuation in the EHF band. This model can estimate the attenuation due to rain up to the frequency of 55 GHz. The parameters used in this model are listed below: R 0.01 : point rainfal l rate for the locatio n for 0.01% of an ave rage year (m m/h) ℎ 𝑠 : height above m ean sea level of the earth station (km) Θ : elevation angle (degre es) 𝜑 : latitude of the earth station (degrees) F : frequency (GHz) 𝑅 𝑒 : effective radius of the earth (8500 km) The geometry is illustrated in Fig. 2. The steps of this model are outlined below: Step 1: Determine the rain height,  ℎ 𝑅 , using the following formula [14] : ℎ 𝑅 = ℎ 𝑜 + 0.36 𝑘𝑚 (1) where ℎ 𝑜 i s t h e 0 𝑜 C isotherm height or f reezing height above the mean sea level (km). The average freezing height for different regions of the world can be found in [14]. Step 2: For 𝜃≥ 5 𝑜 , compute the slant-path length 𝐿 𝑠 , below t he rain height from: 𝐿 𝑠 = ( ℎ 𝑅 −ℎ 𝑠 ) 𝑠𝑖𝑛𝜃 𝑘𝑚 (2) For 𝜃 <5 𝑜 , the following formula is used: 𝐿 𝑠 = 􏷡 ( ℎ 𝑅 −ℎ 𝑠 ) 􏿵 𝑠𝑖𝑛 􏷫 𝜃 + 􏷫 ( ℎ 𝑅 −ℎ 𝑠 ) 𝑅 𝑠 􏿸 􏷪􏷫 ⁄ + 𝑠𝑖𝑛𝜃 𝑘𝑚 ( 3) If ( ℎ 𝑅 −ℎ 𝑠 ) ≤ 0 , th e predicted rain attenuation for any time percentage is zero and t he follo wi ng steps are not required. Steps 3: Calculate the horizontal projection, 𝐿 𝐺 , of t he slant- path lengt h from : 𝐿 𝐺 = 𝐿 𝑠 𝑐𝑜𝑠𝜃 𝑘𝑚 (4) 590 Step 4: Obtain the rai nfall rate, 𝑅 􏷟 . 􏷟􏷠 , exceeded for 0.01% of an average year (w ith an integration ti me of 1 minute). If this lo ng- term statistic cannot be obtained from local data sources, an estimate can be obtained from the maps of rainfall rate g iven i n [10]. If 𝑅 􏷟 . 􏷟􏷠 =0 , the pred icted rain attenuation is zero for any time percenta ge and the f ollowing ste ps are not require d. Step 5: Obtai n the specific attenuatio n, 𝛾 𝑅 , using the frequency- dependent coe fficients given in [1 1] and the rainfall rate , 𝑅 􏷟 . 􏷟􏷠 , determined i n Step 4, by using: 𝛾 𝑅 = 𝑘 ( 𝑅 􏷟 . 􏷟􏷠 ) 𝛼 𝑑𝐵 / 𝑘𝑚 (5) Step 6: Calculate the horizontal reductio n factor, 𝑟 􏷟 . 􏷟􏷠 , for 0.01% of t he time: 𝑟 􏷟 . 􏷟􏷠 = 􏷠 􏷠 + 􏷟 . 􏷦􏷧 􏽰 𝐿 𝐺 𝛾 𝑅 𝑓 −􏷟 . 􏷢􏷧 ( 􏷠−𝑒 􏷫𝐿 𝐺 ) (6) Step 7: Calculate the vertical adjustment factor, 𝑣 􏷟 . 􏷟􏷠 , for 0.01% of the tim e: 𝜉 = 𝑡𝑎𝑛 −􏷠 􏿵 ℎ 𝑅 −ℎ 𝑠 𝐿 𝐺 𝑟 􏷟 . 􏷟􏷠 􏿸 𝑑𝑒𝑔𝑟𝑒𝑒𝑠 For 𝜉 > 𝜃 , 𝐿 𝑅 = 𝐿 𝐺 𝑟 􏷩 . 􏷩􏷪 𝑐𝑜𝑠𝜃 𝑘𝑚 else, 𝐿 𝑅 = ℎ 𝑅 −ℎ 𝑠 𝑠𝑖𝑛𝜃 𝑘𝑚 If 􏿖 𝜑 􏿖 <3 6 𝑜 , 𝜒 =3 6 − 􏿖 𝜑 􏿖 𝑑𝑒𝑔𝑟𝑒𝑒𝑠 else, 𝜒 =0 𝑑𝑒𝑔𝑟𝑒𝑒𝑠 𝑣 􏷟 . 􏷟􏷠 = 1 1+ √ 𝑠𝑖𝑛𝜃 􏿵 31 ( 1 −𝑒 − ( 𝜃 ( 􏷠 + 𝜒 ⁄ ) ) ) √ 𝐿 𝑅 𝛾 𝑅 𝑓 􏷫 − 0.45 􏿸 Step 8: The e ffective path len gth: 𝐿 𝐸 = 𝐿 𝑅 𝑣 􏷟 . 􏷟􏷠 𝑘𝑚 (7) Step 9: The predicted attenuation exceede d for 0.01% of an average year i s given by 𝐴 􏷟 . 􏷟􏷠 = 𝛾 𝑅 𝐿 𝐿 𝑑𝐵 (8) Step 10: The estimated attenuation to be exceeded for o ther percentage of an average year, in the ran ge from 0.001% t o 5%, is determ ined from the atte nuation to be exceeded for 0.01% of an average yea r: If 𝑝≥ 1% or 􏿖 𝜑 􏿖 ≥ 36 𝑜 , 𝛽 =0 If 𝑝 <1 % and 􏿖 𝜑 􏿖 <3 6 𝑜 𝑎𝑛𝑑 𝜃≥ 25 𝑜 , 𝛽 = − 0.005 􏿴 􏿖 𝜑 􏿖 − 36 􏿷 Otherwise; 𝛽 = − 0.005 􏿴 􏿖 𝜑 􏿖 − 36 􏿷 +1 . 8 − 4.25 𝑠𝑖𝑛𝜃 𝐴 𝑝 = 𝐴 􏷟 . 􏷟􏷠 􏿴 𝑝 􏷟 . 􏷟􏷠 􏿷 − ( 􏷟 . 􏷥􏷤􏷤 + 􏷟 . 􏷟􏷢􏷢􏷠 􏸋􏸍 ( 𝑝 ) −􏷟 . 􏷟􏷣􏷤 􏸋􏸍 ( 𝐴 𝑜 . 𝑜􏷪 ) −𝛽 ( 􏷠−𝑝 ) 𝑠𝑖𝑛𝜃 ) 𝑑𝐵 (8) This method provides an estimate o f the long-term statistic s of the attenuation due to rain. IV. R AIN A TTENUAT ION P REDICTION From the previous section, it is clear that the rain attenuatio n is a function of a number of factors , including rain rate, ele vati on angle, frequency, 0 o C isotherm height, height of the earth s t a t i o n f r o m t h e m e a n s e a l e v e l , a n d s o o n . I f a n y o f t h e s e parameters gets changed, the correspondin g rain attenuation will also change. This is w hy determinin g the values of these parameters is a pre requisite for rai n attenuati on predi ction. Considering the longitude (78.5 (-180 o to -180o)) of the satellite TH AICOM-5, the p arameter s for four d ifferent regio ns of Bangladesh are shown in Table II I, wh ere the va lue s of th e first three parameters are taken from [18], and the corresponding antenna elev ation angle for THAICOM- 5 satellite is computed using elevation ang le calculator [19]. Figure-3: Rain Attenuation at K a Band (30 GHz) The 0 o C isotherm height varies signi ficantly from 1 km to 13 km in Bangladesh, However most of the precipitation occurs with freezing height between 3 to 5 km [6]. However, according to ITU-R P.839-3 [14], the average freezing height for the Bangladesh is 4 .5 km. Since the freezing height varies considerably with the season of the year, the value of this parameter is taken from [14]. The regression coefficients, namely k and α , for Ka , Q, and V bands are taken from [15] for simulation. The frequencies of Q and V bands used in simulati on are taken from the U S Federal Communication Communicat ions (FCC)’s proposed frequency bands: 37.5- 40.5 (uplink ) and 47.2-50.2 (dow nlink) for geostatio nary satellite, and 37 .5-38.5 (uplink ) and 48.2-49.2 (downlink) for non-geostationary satellite [16]. The 40 GHz (uplink frequency) of Q band and 50 GHz (downlink frequency) of V band are use d for sim u lation. TABLE III. S IMULATION P ARAMETERS Parameters Dhaka Chittagong Rajshahi Sylhet Longitude (degree) 90.24 91.5 88.36 91.52 Latitude (degree), 𝜑 23.42 22.19 24.22 24.53 height above mean sea level of the earth station (km), h s 4×10 -3 7×10 -3 31×10 -3 9 × 1 0 -3 Antenna Elevation Angle (degree), θ 59.58 60.15 59.59 57.82 Fig. 3 shows the r ain attenuation exceeded at different percentage of time of the year for K a . Th e max imu m ra in attenuatio n of 85 dB is observed for Sylhet, and the minimum of 74 dB for Rajshahi exce eded f o r 0 . 0 1 % t i m e o f t h e y e a r . However, the rain attenuation pr edicted from ITU rain rate is 69 dB exceede d for 0.01% time of the year, which is far lower than t he l owest rain attenuati on obse rved in Rajshahi am ong these four regions from measured rain rate. The rain attenuatio n at Q and V bands a re shown in Fi g. 4 and Fig. 5. Since the rain rate is maximum at Sylhet, the maximum rain attenuations are observed here for both Q and V bands, which are 124 dB and 156 dB, res pectively. On the othe r hand, the minimum rain 591 Fig. 4 Rain attenuation at Q-band (40 GHz) Fig. 5 Rain attenuation at V-band (50 GHz) attenuatio ns of 1 09 dB and 13 8 dB are found in Q and V ba nds for Rajshahi where the rain rate is minimum. But the rain attenuation predicted from the ITU rain rate in Q and V bands are 102 dB and 129 dB, respectively. There is a significant difference in rain attenu ation acro ss Banglad esh. While the maximum attenuation i s obs erved at Sy lhet, the minim um at Rajshahi. The difference in rain attenuation exceeded for 0.01% time of the year between these t w o r e g i o n s i s a b o u t 1 2 % f o r both bands. A significant difference between the rain attenuation predicted from the m easured rain rate and ITU predicted rain rate is found. The rain attenuation pred icted u s ing ITU rain rate data is 6% lower than the least rain attenuation in Bangladesh found in Rajshahi and a bout 18% l ower than the highest rain attenuation in B angladesh observed i n Sylhet. The rain attenuation exceeded for 0 .01% t ime of the year i n V band is observed in Q band for 0.03 % tim e of the year, while the attenuation exceeded for 0.01% time of the year in Q band is same as the 0.001% tim e of the year for K a band. The period of time at which a network remains in outage stag e is an important parameter for determining the quality of the network. T his outage can occur due to the system failure or degradation of sign al quality below a certain threshold. Fo r th e latter one, the signal degradation can occur because of the rai n attenuation. To cope with this, a fade margin of a certain Fig. 6 Average rain fade m argin for 99.99% availability. Fig. 7 Rain fade m argin for 99. 99% availability in different mo nths for Q and V bands. amount is provided. Fig. 6 provides the average fade margin required for 99.99% availability (0.01% outage) of satellite network for Q and V bands. Accor ding to t he c olumn chart, the maximum of 156 dB and 124 dB fade margi n are required in Sylhet region for V and Q bands, respectiv ely, to ensure 99.99% availability, while the m inimum of 138 d B and 109 dB in Rajshah i. However, there is a significan t difference in fade margin predicted from ITU p redicted rain data and the measured rain data. The rain attenuation varies according to the amount of precipitati on. There is a huge variation of p recipitat ion acros s the year. The maximum rain occu rs in July, but the minimum in January. For this reason, the fade margin required to mainta in the certain amount of signal quality v aries over months. Fig. 7 shows how the required fade margin varies over months for 115 120 109 124 102 145 152 138 156 129 0 20 40 60 80 100 120 140 160 180 RainAttenuation( dB) QBand VBand 592 99.99% a vailability in the four regions. The month of July requires the h ighest fade m arg in of a bout 170 dB fo r V ba nd in Sylhet and Chittagong and the sam e for Q band is 145 dB in th e same regions, though the minimum fade margin, which is 40- 60 dB, is requi red in January. The rain fade margin remains more than 140 dB and 1 00 dB for 8 months f rom March to October fo r V band and Q band, respectively. V. C ONCLUSION Attenuation due to rain is a sev ere ba r in d esigning satellite link in frequency more than 10 GHz. There is a large variation in precipitation across Bangladesh and the rain fall rate in monsoon season is 4 to 8 times higher compared to that in dry season. For this reason, the rain attenuation is not same in al l regions of Banglades h, rather it varies from season to season. The rain attenuati ons in Q and V bands are m uch higher (about 32% in Q band and 46% i n V ba nd) than t he existing Ka band. The diffe rence i n the at tenuation across the cou ntry is about 12%, where the maxim um is observed at Sylhet and the least value at Rajshahi region. It is also found that the ITU rain ra te prediction model cannot predict the rain rate properly for a subtropical country l ike Ba ngladesh, because it is found that i ts predicted rain rate is 6% lower than the forty ye ars average ra in rate of the driest region of Bangla desh, and the correspondi ng effect on the rain attenu ation pr edicted from this pred icted ra in rate cannot provide the accurate value. To ensure 99.99% availability of network , the required highest average fade margin of 156 dB and 1 24 dB are reported in Sylhet a nd the minim um values of 138 dB and 109 dB in Rajshahi for V and Q bands, respectively. The required fade margin varies considerably across the year. Th e required margin for 99.99% is more than 140 dB for V band and 100 dB for Q band for about six months. Since providing such a huge link margin is somewhat economically impr a c t i c a l w i t h t h e c u r r e n t technology, the availability will vary with the season in a yea r. During d ry season, highe r availab ility can be enjoyed, but in monsoon th is will go down. However, placing satellite in a lower longitude increases the earth station elev ation angle, which reduces the amount of fad e margin required. In addition, a satellite can be lau nched in wh ich the total number of transpon der will be gro uped int o two cate gories: one operating in lower frequency band and t he other in Q/ V band. During the monsoon seaso n, the lower frequency b and will be u sed, and the EHF band will be in operating mode in the d ry season. Transmitting higher power du ring rain and reducing that in other time can a lso reduce the effect of rain in th e quality of satellite services. R EFERENCES [1] E. Cianca , T. Rossi, A. Y ahalom, Y. Pinhasi, J. Far sero tu and C. Sacchi, “ EHF for Satellite Communications: the New Broadband Frontier ”, in Proc. of the IEE E, vol. 99, no. 11, Nov. 2011. [2 ] J. D. La st e r, W . L. St ut zm an , “ Frequency Scaling of Rain Attenuation for Satellite Communication Links ”, IEEE T rans. on Antennas and Propagation, vol. 43, no. 11, pp. 1207 -1216, 1995 [3] M. M. Rashid, M. R. Islam, AHM Z. A lam, S. Kh an, O. O. Kh alifa and M. A . R a h m a n , “ Rain Fade and It’s Effect A nalysis for Earth-to-Satellite Microwave Link Based on Measured Rainfall Statistics in Bangladesh ” ,in Proc. 2009 IEEE Malaysian Int’l Conf. on Communication, pp. 244- 249. [4] S. Ho ssa in, “ Rain attenuation prediction for terrestrial microwave link in Bangladesh ”, Journal in Electrical and Electronic Engineering, vol. 7, no . 1, pp. 63-68, 2014. [ 5 ] M . S . H o s s a i n a n d M . A . S a m a d , “ The tropospheric scintillation prediction of earth-to-satellite link for Bangladeshi climatic condition ”, Serbian Journal of Electrical E ngineering, vol. 12, no. 3, pp. 265-273, 2015. [6] M. N. Isla m, H. Uyeda, “ An analysis to classify pre-monsoon and monsoon rainfall in and around Bangladesh us ing TRMM 2A25 data ”, Autum n meeting of the m eteoro logical society of Japan, pp. 1 4-16 October 2007, H okkaidoo, Japan. [ 7 ] M . K a m r u z z a m a n a n d M . S . I s l a m , “ Rain Attenuation Prediction for Satellite Communications link at Ku and Ka bands Over Bangladesh ”, in Pr oc. 2014 IEEE I nt’l Conf. on Electrical Engineering and In formation & Communication T echnology (ICEEICT) , pp. 1-6. [8] Bangladesh Agricu lture Research Council, http://www.barc.gov.bd/ym_rainf all.php , 03 March, 2014. [ 9 ] P . L . R ic e a n d N . R . H o l m be r g, “ Cumulative Time Statistics of Surface- Point Rainfall Rates ”, IEEE Trans. on Communications, vol. 21, no. 10, pp. 1131-1136, 1973. [10] Recomm endation IT U-R P.837-6, “ Characteristics of Precipitation for Propagation Modelling ”, Geneva, 2013 [11] G. S. Feldhake and L. A. Sengers, “ Comparison of Multiple Rain Attenuation Models with Three Years of Ka Band Propagation Data Concurrently Taken at Eight Different Locations ”, Journal of Space Communication, vol. 2, Fall 2002. [12] Recomm endation ITU-R P.618-10 (10/2009), “ Propagation data and prediction methods required fo r the design of earth- space telecommunication System ”, P Series. [13] CCIR Report 564-3, 1986. [14] ITU -R P.839-3, “ Rain height model for prediction methods ”. 2001. [15] I TU-R P.838-3, “ Specific attenuation model for rain for use in predict ion methods ”, 2005. [16] J. V. Evans and A. Dissanayake, “ Prospects for Commercial Satellite Services at Q-and V-bands ”, in Proc. 1998 IEEE Conf. on Military Communications, pp. 1-7. [17] Dennis Roddy, “Satellite Com munications”, Fourth Edition, The McGraw-Hill Companies, Inc, 2006. [18] Day and time, http://datea ndtime.info , 03 March 2014. [19] Thaicom, http://tcns.thaic om.net/AZEL.as p , 03 March 2014 . 593