SpaceOps Conferences 16-20 May 2016, Daejeon, Korea SpaceOps 2016 Conference 10.2514/6.2016-2434 A Case Study of the Data Downlink Methodology for Earth Observation Satellite Akio Oniyama 1 and Tetsuo Fukunaga 2 PASCO CORPORATION 4-10-1 Nakano, Nakano-ku, Tokyo, Japan +81 3 5318 1084 For the Earth Observation Satellites, the demand to expand high-resolution imaging for wide-swath has been gradually increased in the past recent years. Meanwhile, frequency allocation for data downlink is X band and band width is limited to only 400MHz. The current paper describes the results of a case study in order to obtain wider band downlink using Ka band. Keywords: Space Communication, Earth Observation Satellite, Ka band, Rain Loss, Atmospheric Loss 1. Introduction The current X band direct downlink is possible to reach maximum data rate of about 800Mbps using 16QAM. However, presently the demands are for a higher resolution and a wide-swath, and we must need a further expansion of the data downlink rate. The downlink rate can be doubled using the dual polarization transmission. If we use the multi-level modulation such as 16QAM and higher level modulation by utilizing dual polarization, it may cause some problems for instance transmission quality degradation due to cross polarization interference. To meet the increased demands, if we use higher frequency, Ka or V band, we can get wider frequency band than X band. In the United States and Europe, there is a start of the usage of Ka band or Optical downlink carrier. There is a plan to use Ka band as one of the solutions for increasing the data rate of Earth observation satellite in Japan. We know very well that there is increase in the "rain loss" and "air loss" etc., when we use higher frequency. Japan is located in the eastern part of Asia where it has a rainy and heavy monsoon region with average annual rainfall of 1,718mm. If we use Ka band for the communication, the "rain attenuation" has to be considered as an essential factor. Our company s business model is to collect the data from the Earth Observation Satellites and to generate value added products. PASCO is a leading geospatial company and provides a consistent service from the acquisition, to analysis and provision of geospatial information to the customers in Japan. We perform satellite operations in order to provide higher customer satisfactions. There are not many private companies conducting such kind of the satellite business in the world. We made trade off regarding some of the impacts when we use Ka band for our Earth Station Network in Japan. 2. Estimation of Propagation Loss defined by ITU We have two Ground Stations in Japan located in Hokkaido and Okinawa prefectures. We are operating every day the Earth Observation Satellite using our Ground Stations. We use S band and X band to communicate with the satellite. However, the needs for higher frequency are increasing due to increasing resolution and wide-swath. In Japan, the JAXA has a plan to use Ka band for data downlink in the near future. We are realizing that these needs are gradually increasing, therefore we have to prepare for these growing needs in the near future. We estimate communication condition using Ka band for our satellite network according to ITU recommendation. Fig 2-1 shows location of our Ground Stations and Table 2-1 shows the Ka band specification under planned. Table 2-2 shows the list of propagation loss defined by ITU-R618-12. We calculate propagation loss of our Ground Station according to this recommendation after paragraph 2.1. 1 Technical Adviser, Satellite Business Devision, aakmia9820@pasco.co.jp 2 Dupty General Maneger, Satellite Business Devision, taegta9186@pasco.co.jp 1 Copyright 2016 by the, Inc. All rights reserved.
Table 2-1 Ka band specification Fig. 2-1 Location of PASCO s Ground Station Table 2-2 List of propagation loss on an Earth-space path Item Effects for Ka band Detail Rec. Attenuation by atmospheric gases To increase with frequency above 10 GHz, especially for low elevation angles. ITU R P.676 Attenuation by rain, other precipitation and clouds Clear-air effects Beam spreading loss Decrease in antenna gain due to wave-front incoherence Scintillation and multipath effects The largest raindrops 5mm. The wave length of 10GHz is 3cm. The frequency higher than 10GHz is greatly affected by the absorption and scattering by rain. Frequency below about 10 GHz and at elevation angles above 10. At low elevation angles ( 10 ) and at frequencies above about 10 GHz, however, tropospheric scintillations can on occasion cause serious degradations in performance. Can be ignored at elevation angles above about 3 at latitudes less than 53 and above about 6 at higher latitudes. This effect increases both with increasing frequency and decreasing elevation angle, and is a function of antenna diameter. Although not explicitly accounted for in the refraction models presented below, this effect is negligible in comparison. Amplitude scintillations increase with frequency and with the path length, and decrease as the antenna beam width. Attenuation by sand and dust storms These attenuations become particularly severe at frequencies higher than 10 GHz, especially for very small aperture terminal Noise temperature T sky = T mr (1 10 A/10 ) + 2.7 10 A/10 K T mr = 37.34 + 0.81 T S K A :total atmospheric attenuation db ITU R P.838-3 ITU-R P.834 ITU-R P.834 ITU-R P.834, NA ITU R P.372 Cross-polarization effect Propagation Loss 1 of rotation may be encountered at 10 GHz, and greater rotations at lower frequencies. 32.45+20log(1/λ)+20log(d) λ:wave length d:slant range ITU R P.531 NA 2
2.1 Attenuation by atmospheric gases Fig.2-2 shows the characteristics of Dry air and Water vapor defined by ITU-R P.676. Attenuation at 26GHz due to the atmosphere gases is dominated by absorption and scattering of water vapor, value assumes 1dB. The relationship between the antenna elevation angle (EL: θ) and this attenuation is calculated by the following equation. A=hw γw/sinθ Where hw is 10km, γw is 0.1dB/km. Fig. 2.-3 shows the relationship between altitude and density of water vapor. We can read out nearly equal 0dB at higher than 10km from fig. 2-3. Fig. 2-4 shows result of this calculation. This shows 11.5dB attenusation at EL=5 which is a large value. It may lead too difficult to demodulate correctly. We need some specific countermeasure at our Ground Station for this effect. Fig. 2-2 Specific attenuation of atmospheric gases Fig. 2-3 Vertical distribution of water vapor Fig. 2-4 Attenuation due to water vapor 3
2.2 Attenuation by rain, other precipitation and clouds Fig.2-5 shows the results of estimation on relationship between Rain rate and Attenuation. This figure shows both of attenuation X band and Ka band. The Ka band attenuation of km path length is 2 digits higher than X band attenuation. Table 2-3 lists the maximum rain rate at Hokkaido and Okinawa Ground Stations from 2003 to 2015. Communication path is 10 minutes or so for the Earth Observation satellite. The value of 10 minutes rain rate is 19mm for Hokkaido and 2.35mm for Okinawa. We can calculate 3dB/km for Hokkaido and 3.8dB/km for Okinawa as maximum attenuation level. We estimated a relationship between antenna EL and attenuation level (Fig. 2-5). These values suggest that it is too difficult to demodulate accurately under this rain condition, if we use larger antenna. However, the attenuation level at 1mm rain rate is 9dB when antenna El angle is 5 deg. This level may be still too big to demodulate signal accurately at near AOS or LOS. But, we can conclude that it is not a problem to operate Ka band downlink satellite because the possibility of rain rate more than 1mm per 10 minutes is 9% in each year. Further, in Okinawa we have the possibility of more than 1mm rain rate per 10 minutes as 15% per year. Fig. 2-5 Attenuation per km due to Rain Table 2-3 Actual Record of Rain Rate at Hokkaido Ground Station 4
Table 2-4 Actual Record of Rain Rate at Okinawa Ground Station Table 2-5 Attenuation due to Rain 5
Fig. 2-6 Attenuation due to Rain 2.3 Clear-air effects In ITU R-618, the following statements are described. Clear-air effects produce serious fading in space telecommunication systems operating at frequencies below about 10 GHz and at elevation angles above 10. At low elevation angles (< 10 ) and at frequencies above about 10 GHz, however, tropospheric scintillations can on occasion cause serious degradations in performance. Ka band may receive some degradation at antenna EL angle at less than 10deg as tropospheric scintillations. We have mentioned about the estimation of this degradation in section 2.6. 2.4 Beam spreading loss The regular decrease of refractive index with height causes ray-bending and hence a defocusing effect at low angles of elevation (recommendation ITU-R P.834). The magnitude of the defocusing loss of the antenna beam is independent of frequency, over the range of 1-100 GHz. The loss Abs due to beam spreading in regular refractive conditions can be ignored at elevation angles above about 3 at latitudes less than 53 and above about 6 at higher latitudes. At all latitudes, the beam spreading loss in the average year at elevation angles less than 5 is estimated from: Abs = 2.27 1.16 log (1 + θ) db θis EL angle. Table 2-4 Beam Spreading Loss due to Defocusing This shows level is very low, so this effect is negligible. 6
2.5 Decrease in antenna gain due to wave-front incoherence In ITU R-618, the following statements are described. This effect increases both with increasing frequency and decreasing elevation angle, and is a function of antenna diameter. Although not explicitly accounted for in the refraction models presented below, this effect is negligible in comparison. 2.6 Scintillation and multipath effects In ITU R-618, the following statements are described. The amplitude of tropospheric scintillations depends on the magnitude and structure of the refractive index variations along the propagation path. Amplitude scintillations increase with frequency and with the path length, and decrease as the antenna beam width decreases due to aperture averaging. Fig. 2-7 shows the level of fading depth due to tropospheric scintillations defined by ITU-P618. It means there is 5 db at EL 5 deg. and increasing fading level <5 deg. It is not so high level to operate ground station. Usually, Ka band communication begin more than 5 deg antenna EL angle. Fig. 2-7 fading depth due to tropospheric scintillations 2.7 Attenuation by sand and dust storms In ITU R-618, the following statements are described. Very little is known about the effects of sand and dust storms on radio signals on slant-paths. Available data indicate that at frequencies below 30 GHz, high particle concentrations and/or high moisture contents are required to produce significant propagation effects. We face seasonal sand and dust storms from China. Table 2-5 shows the average days observed yellow sand from 1981 to 2010 by JMA. There is no data on sand volume. So, we cannot estimate the impacts due to these sand storms. According to this data, there are lots of sand storm from Mar to May during spring season. We are operating X band downlink at our Ground Station. However, we have no report from degradation due to sand storm. But, there is some impact due to this sand storm for Ka band operation. There are few published papers about the impacts due to sand and dust in desert areas, and therefore, we will study and estimate our impacts in near future. Table 2-5 Average Days of Sand storm from China Jan Feb Mar Apr May Jun July Aug Sep Oct Nov Dec total Days 0.5 2.2 6.9 9 4.1 0.4 0 0 0 0.2 0.5 0.5 24.2 Ave./month 1.6% 7.9% 22.3% 30.0% 13.2% 1.3% 0.0% 0.0% 0.0% 0.6% 1.7% 1.6% 6.6% 2.8 Noise temperature The sky noise temperature at a ground station antenna may be estimated by: Tsky = Tmr (1 10 A/10 ) + 2.7 10 A/10 7
where: Tsky : sky noise temperature (K) at the ground station antenna A : total atmospheric attenuation excluding scintillation fading (db) Tmr : atmospheric mean radiating temperature (K). When the surface temperature Ts (K) is known, the mean radiating temperature, Tmr, may be estimated for clear and cloudy weather as: Tmr = 37.34 + 0.81 T S K Fig. 2-7 Noise Temperature due to Atmospheric gasses 2.9 Cross-polarization effect This effect can be estimated using the method described ITU-P618. Step 1 C f 60 log f 28. 3 26 log f 4. 1 35. 9log f 11. 3 Step 2 C A = V ( f ) log A p A p :rain attenuation (db) 6 f 9 GHz 9 f 36 GHz 36 f 55 GHz Step 3 0.21 30.8 f 6 f 9 GHz 0.19 12.8 f 9 f 20 GHz V f 22.6 20 f 40 GHz 0.15 13.0 f 40 f 55 GHz Step 4 C = 40 log (COS ) for 60 Step 5 C = 0.0053 2 takes the value 0, 5, 10 and 15 for 1%, 0.1%, 0.01% and 0.001% of the time, respectively. Step 6 XPD rain = C f C A + C + C + C Fig. 2-8 shows relationship between Rain rate and XPD level at EL 5 deg. We can read out 40dB at 10mm rain rate from Fig.2-8. This value is 10dB lower than X band XPD at the same condition. In Hokkaido, the percentage that we have the rain more than 10mm/h is about 7% in each year. In Okinawa, it is about 6%. This means it is not so severe condition for our Ground Station. 8
Fig. 2-8 XPDrain due to Rain 2.10 Propagation Loss We compared propagation loss for S band. The behavior of X band and Ka band are shown in Fig 2-9. We can know 10dB as higher loss than X band. Fortunately, the antenna gain improved about 10dB in comparison with X band if we use the same size antenna diameter. So, this means it is not so severe condition. Fig. 2-9 Propagation Loss for S, X Ka band 3. Conclusion Table 3-1 shows the results summary of this case study. We are now using 7.3m Antenna. We can use the current antenna to receive Ka band signal in near future. However, there are some impacts in order to operate with the same as operation quality when we will use the current antenna. There are some problems when we operate at AOS and LOS. We may receive some anomalies due to Atmospheric gases, Rain, and increasing of Noise Temperature. We may need to solve these problems. However, it is not so difficult issue. It may be able to solve to use bigger size dish antenna. Anyway, this paper is just our study, we need to think more in order to operate Ka band signal at our ground station. We will continually research and study. 9
Table 3-1 results summary of this case study Item Attenuation db Remarks Attenuation by atmospheric gases 11.5dB at EL=5 deg. We need some countermeasure Attenuation by rain, other precipitation and clouds 8.7dB at EL=5 deg. under 1mm/h We need bigger antenna than X band Clear-air effects NA To be considered tropospheric scintillations Beam spreading loss Decrease in antenna gain due to wavefront incoherence 0.01 db at EL=5 deg. NA negligible negligible Scintillation and multipath effects 5dB at EL= 5 It Is not so high level. deg. Attenuation by sand and dust storms NA May be some impact to operation. TBD Noise temperature >250 K at EL=5 deg. This is about 160K bigger than X band. We need some countermeasure Cross-polarization effect 50dB at 1 mm It is not so severe condition rain Propagation Loss 10dB higher than X band It can compensate Antenna gain. 4. References (1) ITU R P618-12 Propagation data and prediction methods required for the design of Earth-space telecommunication systems (2) ITU R P-838-3 Specific attenuation model for rain for use in prediction methods (3) ITU R P-834-5 Effects of tropospheric refraction on radio wave propagation (4) ITU R P-374-12 Radio noise (5) ITU R P-531-12 Ionospheric propagation data and prediction methods required for the design of satellite services and systems (6) Systems Adaptation for Satellite Signal under Dust, Sand and Gaseous Attenuations Kamal Harb*, Omair Butt, Samir Abdul-Jauwad, Abdulaziz M. Al-Yami Electrical Engineering Department, KFUPM University, Dhahran 31261, Saudi Arabia Journal of Wireless Networking and Communications 2013, 3(3): 39-49 (7) Estimation of croos polarization due to rain over some station in India R Sen Jaiswal, P Geetha & S Uma Department of Physics, Sona College of Technology Iandian Jpurnal of radio & Space Physics Vol. 36, Oct. 2007 pp.379-382 10