SATCOM Final Report Arnaud Dannequin

Arnaud DANNEQUIN

THE VOLVO OCEAN RACE 2015

RRY100

Titles you can't find anywhere else

Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.









Arnaud DANNEQUIN

RRY100

SUMMARY Abstract Introduction I. Satellite specifications: “Climb out and head to space” I.1 2015 issue selected route I.2 Orbits I.3 Footprints I.4 Description of the overall communication system I.5 On-board satellite antennas and electronical components I.6 Satellite system noise temperature II.Ship specifications: « Between sky and sea » 2.1 On-board ship antenna 2.2 Ship receiver/transmitter system 2.3 Ship system noise temperature III. Ground station: « Down to earth » 3.1 Earth station antenna design 3.2 Earth station transmitter-receiver system 3.3 Earth station system noise temperature IV Signal propagation: « Every cloud has a silver lining » 4.1 Atmosphere attenuation 4.2 Rain attenuation 4.3 Galactic noise 4.4 Sun interferences 4.5 Free space loss 4.6 Frequencies, modulation and bandwidth 4.7 Multiplexing and polarization V Link budget: « Receiving you load and clear » 5.1 Earth station to GEO satellite communications 5.2 Earth station to LEO satellite communications 5.3 Ship to LEO satellite communications 5.4 LEO to GEO communications 5.5 Review of the obtained results









Arnaud DANNEQUIN

RRY100

Abstract This report proposes a possible model to provide satellite communication for the 2015 Volvo Ocean Race. It deals with the main parties: technical description of chosen satellites (orbits, antennas, electrical components), earth station (antenna, transmitter/receiver system) and on-board ship communication system. The purpose of this paper is to verify communication link quality of this system by fixing transmitting powers, the data rate in order to find the Bit Error Rate, best parameter for describing communication quality. Calculations are based on the “worst case method”. In the other hand, the considered attenuations are also mentioned. The conclusions of this report are that link qualities are, other than GEO-earth station links, very -4 -15 acceptable in view of extreme weather conditions. Indeed, BER range between 10 and 10 . Key words: satellite, earth station, ship, link budge t calculation

Introduction The Volvo Ocean Race is one of the most prestigious world sailing races since 1973. For the 2015 issue, 20 pairs of boats will start the race at Gothenburg. After 9 challenging stages taking place in all oceans of the world, they will return hopefully, to their starting point. From telecommunications point of view, each crew must be able to communicate with their other team crew, their team bases, organizers, and media such as journalists. More technically, as a consultant in the Satellite Communications department, I have to design a telecommunication system, which allows voice communications, HDTV, and all data needed relating to the sailing navigation taken from inboard sensors. The required data rate must be equal at least to 2MB/s. Furthermore, the designed telecommunication system in the aim of transferring data, must be available most of the time (99.999 % of the time). We assume that infrastructure needed such as the draft budget and the issues relating to the launch of our used satellites will not get expanded in this report.

I. Satellite specifications: specifications: “climb out and head to space” I.1 2015 issue selected route The route selected by the organizers for the 2015 edition is the same, as the previous issue except the start is not based in Alicante but in Gothenburg. This route is the starting point to know which areas we have to cover and those we have to ignore as the North and South Pole.









Arnaud DANNEQUIN

RRY100

1.2 Orbits The first idea was to use only GEO satellites with several earth stations. In that way, this simplifies the project. However, because GEO are in orbit 35784 km away from Earth, the delay of voice communications is around 250 ms (around 10 ms for LEO satellite). That is why I chose to combine LEO satellites with one GEO satellite. The GEO satellite is pointing at Gothenburg. Because its revolution is the same as Earth its always visible from it. To cover almost the entire globe I choose an existing LEO constellation called Globalstar constellation. It is composed of 48 LEO satellites and 4 spare satellites for redundancy. The inclination of their orbits its 52° degrees, 1414km away from earth with a velocity around 3km/s. We consider that this satellite constellation works for both KU-band and KA-band (inter-satellites communications). Since, the global star constellation doesn´t cover all ocean regions, I decided to be sure, to put on orbit 10 more satellites. The other advantage of the LEO satellites is because they are closer to Earth, the free space loss is lower than for GEO satellites. On the other hand, their revolution period is very short and they are mobiles, that makes the design much more difficult and increases the number of satellites required in order to cover the globe.

Globalstar constellation

Because “All roads lead to Rome”, I could use another communication system. I could use one GEO pointing at Gothenburg and a constellation of MEO orbiting around the earth. But the main disadvantage is that the delay for voice communications is higher than a communication system based on LEO constellation; since Medium Earth Orbit satellites are orbiting around the earth situated from 5000 km to 12000 km.

1.3 Footprints On the figure of the global satellite footprints of the Globalstar constellation we see that the North Pole and South Pole are not covered. This is not a problem since the boat won´t navigate in these areas. However, we can see that the Globalstar satellites do not cover some ocean regions. With our 10 more satellites we will be able to cover all ocean regions.









Arnaud DANNEQUIN

RRY100

1.4 Description of the overall communication system

Legend: link 1: Voice + HDTV in Ku-band; link 2 : HDTV in Ka-band; link 3: Voice in Ku-band link 4: HDTV in Ku-band Voice communication system (ship to ship communication)

LEO

SHIP

SHIP

I use a LEO satellite in order to provide voice communications between ships. The LEO operates as a relay of the communication for handover management. Moreover, if the two ships are very distant between them, I will use a second LEO to provide the call. The first LEO will send the incoming call to the second and the latter will send it to the recipient. This system is called inter-satellite handover. Ship - home harbour communications After receiving data from the ships, the earth station b ased in Gothenburg sends the data to the respective home harbours by using optical fibber links. Moreover, optical fibber technology is very expensive but we have an unlimited budget. The advantage is that the data rate is high (around 100 Mb per second). HDTV and inter-satellite communications After receiving the signal from the boat, the LEO demo dulated it and sends only HDTV data to the GEO, which send finally to the earth station. In the case that the GEO is not visible from the LEO, which received the signal from the ship, we use the other LEO or several LEO as relays to reach it.

1.5 On-board satellite antennas and electronical components On-board L.E.O satellite antennas For providing ship-LEO and earth station-LEO communications I chose to use two reflector antennas (one as transmitter and one as receiver), which works for a frequency of 12 GHz (Ku-Band).









Arnaud DANNEQUIN

RRY100

RUAG reflector antennas

To ensure inter-satellite communications (both LEO-LEO communications and GEO-LEO communications) we will use also reflector antennas (one as receiver and the other as the transmitter). Because I decided to choose a 30 GHz the antenna diameter will be smaller than Ku-Band: To have a reasonable gain I fixed a diameter of 20 centimetres and an aperture efficiency of 0.8 Thus, the gain of the antennas for inter-satellites communications becomes: 2

G=! ("d/#) = 35 [dBi] I fixed the transmitted power at 100 W. During my visit in RUAG, I learned by the antenna department director that fibber carbon is used for making satellite antennas because fibber carbon doesn´t have any distortion features. It is quite interesting to use it because one side of the dish is in the darkness (cool temperature) and the other side is enlightened by the sun (hot temperature). Therefore, all of my spacecraft reflector antennas are made in fibber carbon. On-board LEO receiver system design The receiving LEO satellite chain system is quite different from the earth station receiving system. It is composed of: One feeder I chose the same feeder as the earth station. Thus, LF =2 [dB] ,T F =290 [K] and T efeeder = 106 [K] •

One demultiplexer Since the received signal is composed of several data, we need a demultiplexer placed after the feeder in order to separate the users, It will allows LEO satellites to send the right data to the right user. •

One Traveling Wave Tube (TWT) A TWT amplifies the power of the weak signal from earth . I chose VZU-6992EB TWT model as the T WT (from Teledyne datasheet) with: FTWT=5 [dB] (fixed 10dB max.) GTWT =70 [dB] Effective temperature: T eTWT=T0(FTWT-1)=290(3.16-1)=626 [K] We consider that this TWT works both for KU-band and KA-band with an almost identical noise figure. •

One Ku-band receiver I chose current generation Telecom 2D type as the Ku-band receiver (datasheet from ping pong p.23) : Noise figure F r = 2 [dB] Gain =63 dB effective temperature : T er =T0(Fr -1)=290(1.58-1)=168 [K] •











Arnaud DANNEQUIN

RRY100

effective temperature : T er =T0(Fr -1)=290(1.7-1)=203 [K] Transponder (or repeater) We will use for transmitting the signal a transponder system. The transponder has to demodulate and remodulate the signal. Indeed, we use a modulation for the received signal to fit the transmission channel. We could use a “regenerative transponder” in order to do the all processing signal (filtering, amplification, demodulation etc). I won´t describe it because it´s quite complex. There are several advantages to use a regenerative repeater. Indeed, it can increases channel capacity, and minimize the interferences. On-board GEO satellite antennas

RUAG horn antenna

For global coverage purpose, horn antennas are commonly used. Since we want a wide coverage beam for our geostationary satellite I chose this type of antenna. We will use one as transmitter and one as receiver. I decided to take the RUAG’s model of KU-Band horn antenna. On RUAG’s datasheet it’s mentioned that the gain of the antenna is 17dBi.I fixed the aperture efficiency to 0.8 to minimize side lobes and since I didn’t find it on the RUAG’s datasheet. So the diameter d of the conical horn antenna is: 2

G=! (("d/#)) =17dBi Finally, d= (#/")$(G/!) = 3.7 [cm]

! : efficiency factor of 0.8 # : wavelength of 0.025 m G : gain of the horn antenna Since the horn antenna’s power is not provided by RUAG i chose to fix it. I fixed the transmitting power of the antenna at:









Arnaud DANNEQUIN

RRY100

1.6 Satellite system noise temperature GEO and LEO system noise temperature Since GEO and LEO have the same main components with the same noise temperatures, these two kinds of satellites have the same system noise temperature. We consider that the noise antenna temperature is about 290 K. Thus, the effective noise temperature of the antenna: Teantenna =Tantenna/Lfeeder =290/1.58 =183 [K] The system noise temperature is: Tsat.system =Tin +TeRX = Teantenna + Tefeeder +TTWT =230 + 106 + 626 = 915 [K]

II. Ship specifications: « Between sky and sea » 3.1 On-board ship antenna For my antenna, I chose a Ku-Band maritime reflector antenna with full 360 degree turning ability and with an elevation angle ranged between -25° and 125°. From the diameter in the datasheet the gain of the on-board ship antenna is: 2

-2

2

G=! (("D/#)) =0.8*(("*130*10 /0.025)) =43 [dBi] The other characteristic we need it’s the transmitting power of the antenna given by the constructor: Pt = 175 W

3.2 Ship receiver/transmitter system We will use as our receiver system for the ships the same system as the earth station with same component characteristics. However, each ship has to send voice and HDTV. Thus, we need to add a multiplexer to send as the same time into a single signal voice and HDTV. Then, we placed the multiplexer after the feeder.

3.3 Ship system noise temperature I consider that we have the same components (feeder, LNA and IF amplifier) with the same parameters on ships. Thus, ship system noise temperature is equal to the earth station system noise temperature:









Arnaud DANNEQUIN

RRY100

Then we can calculate the half-power beamwidth (HPBW) :

%3dB=70°*(c/f*D) =70*((3.108)/(12*109*20)) =8.75*10-2° After, I fixed the transmitted power of the antenna : Pt = 200 [W] Then, I can calculate the EIRP of the antenna: EIRP = Pt +Gt = 23 [dBW] + 67 [dBi]= 90 [dBW]

3.2 Earth station Transmitter-receiver system design For both receiver and transmitter block we need a duplexer as a switch, which allows the antenna to both works in receiving mode and transmitting mode. Thus, we need only one antenna on the earth station. We place it after the feeder at the beginning of the receiver system and at the end of the transmitter system. Earth station receiver system I chose a simple receiving system for the earth station composed of: •

One horn as feeder and waveguide

After the signal was reflected on the antenna’s dish (concentrated o n the focus), to collect the whole waves and to transport them to the receiver system to tune them. We put it close to the antenna and close to the receiver system to minimize ohmic losses. I chose a usual value of LF =2 [dB] and T F =290 [K]

•

One Low Noise Amplifier (LNA)

The main reason to use a LNA is the signal has travel a long way to reach the ground station. At the end of its path, the signal is very weak. Therefore, the LNA amplifies it without adding a high noise. As the LNA I used a FET cooled Peltier (for f=12 GHz TLNA=120 K, GLNA =60dB) because it has high gain and noise temperature relatively low. Indeed, the most important component in the receiving system is the first one it should have a high gain and low effective input noise temperature. The parameters of the other components are less important. •

One down converter (DC)

Since electronic devices such as oscilloscope or computers don’t work for high frequency we need to lower the frequency of the input signal. Moreover, the demodulation needs to be in low frequencies. For this we commonly add a down converter which it lowers the signal to allow engineers to do measurements or to









Arnaud DANNEQUIN

RRY100

Earth station receiver block diagram •

One demodulator In the aim of recovering the envelope of the modulated signal to separate the voice and the HDTD data we need a QPSK demodulator. The demodulator is placed at the output of the receiver system. Earth station transmitter system The earth station transmitter block is as follow:

QPSK modulator

UP converter

IF amplifier

Duplexer

3.3 Earth station noise temperature Calculation of the system noise temperature First we need to calculate the noise temperature from the receiver system. The first element (LNA) is the most important component for the system noise contribution of the receiver system. We can neglect the contribution from both the down convertor and the IF amplifier because the contribution is less and less significant. TRX &TLNA=120 [K] We consider the worst case with heavy rain conditions and an elevation angle of 2° (longest path) Tantenna= Tsky/Arain + Tm (1-1/Arain) + Tground for an elevation angle of 2° : T ground = 50 [K] and heavy rain conditions (worst case) increase the sky noise temperature around 250K ( picked up from homework 5) for an elevation angle of 2°. Tsky =200 [K] in heavy rain conditions T m =290 [K]









Arnaud DANNEQUIN

RRY100

Atmospheric gases especially oxygen and vapour water absorb energy of electromagnetic waves. From the figure below, we see that attenuation from these two gases is very important for a frequency ranged in 22 GHz to 65GHz.SInce our chosen frequency is 12 GHz the resulting attenuation is usually around less than 1 dB (closer to 0.001 dB for our case). This is precisely what has incited me to neglect atmospheric gases attenuation. attenuation .

Atmospheric Atmosph eric gase s attenuation attenua tion in fun ction of fre quency

Atmospheric cloud attenuation For modelling atmospheric cloud attenuation I chose the Rayleigh approximation that applies for frequencies below 200 GHz:

'c = Ki*M [dB/km] where: 'c: specific attenuation (dB/km) within the cloud Ki: specific attenuation coefficient ((dB/km)/(g/m3)) M: liquid water density in the cloud or fog (g/m3) (M=0.5 [g/m3] worst case of a thick cloud By analysing the ITU’s figure of specific attenuation by water droplets at various temperatures as function of frequency, I found: Ki & 0.15 [(dB/km)/(g/m3)] Finally, 'c = KiM = 0.075 [dB/km] Then, for the next calculation, I took the value of the tallest cloud ever recorded (14 miles &22.5 km) thunderstorm. Finally, the worst cloud attenuation is:











Arnaud DANNEQUIN

RRY100

I used the worst-case method. I took the minimum elevation angle to take the longest path length to do my calculations. That is why i took an elevation angle equal to 2°. b

Arain = a* R *Leff For a frequency of 12 GHz the two parameters are: -5

2.42

a=4.21*10 *f

-0.0779

=0.017 and b= 1.41*f

= 1.162

As mentioned previously, the rain rate R [mm/h] is taken from the ITU -R table: R=250 mm/h The path length L rain is equal to: Lrain=(He-H0)/sin(()=100 [km] He : effective rain rate Hi : zero degree isotherm H0 : difference between the earth station’s altitude and the sea level (H 0=10 meters for Gothenburg) ( : elevation angle ( ( =2° worst case) For R> 10 mm/h and |A e|>30° : He=Hi+log(R/10)=3.45 [km] where Hi=7.8-0.1*|A e|=2.057 [km] Ae is the latitude of the earth station (based in Gothenburg equal to : 57°43'N, 11°59'E Lrain=(He-0)/sin( ()=98.6 [km] Then, f=12 [GHz] ) =1/22 (empirical quantity) Leff =((1-exp(-)*b*log(R/10)*Lrain*cos(()))/()*b*log(R/10)*cos( ()))=13.54 km Finally,









Arnaud DANNEQUIN

RRY100

4.4 Sun interference Sun interferences could occur when the earth station is pointing at the satellite and the sun at the same time. Thus, the antenna is duped by the fact that the energy from the sun and the signal satellite are combined . 4

Since %3dB < 0.5° according to Lecture 3 formula and figure, for a frequency of 12 GHz: *T A =TSun&10 [K] The configuration of the earth station antenna pointing at the satellite and the sun situated just behind it occurs only few days per year and only few minutes during these days. Moreover, sun outage doesn’t affect LEO; only GEO satellites because of their fixed position relative to earth. Therefore, I won’t take consideration of sun interferences for my calculation. Probably, my communication system won’t work during these short periods. However, I calculated the system availability according to Lecture 2 : SysAv = (required time – down time) / (required time) with required time: 1 year and down time : 6*8 minutes Then, SysAv =99.9985&99.999% My system availability is very close to the required system availability even the sun affects the communications.

4.5 Free space loss Free space loss for earth station/ship – LEO link LFS= (4*"*R/#)^2 =177 [dB] where R : distance from the LEO satellite to earth Free space loss for earth station – GEO link LFS = (4*"*R/#)^2 =205 [dB] Free space loss for earth LEO – GEO link









Arnaud DANNEQUIN •

RRY100

3 MHz margin losses such as Doppler effect

Moreover, between each channel there is an unused bandwidth called guard band of 5MHz for preventing interferences. For the 2015 issue, there will be at the start 40 boats. Moreover, we use a QPSK modulation. 6 6 Thus, the required bandwidth B is: B=(8*10 *40+5*10 *40)/2=260 MHz

4.7 Multiplexing and polarization Multiplexing The purpose of multiplexing is to share a channel between several users. Choosing the type of Multiplexing that is choosing the way we share the bandwidth between the users. I chose Time-division multiplexing (TDM) (user bandwidths are separated in time but they have the same frequency) instead of FDM (Frequency-Division Multiplexing) (Each user bandwidth is separated in frequency) because HDTV is a digital technology and need much more bandwidth than voice, which is an analog signal, but by using an analog-digital convertor we could use TDM for both. Polarization: Using different polarizations for each link protect the reception of signals of identical or closely spaced frequencies, since it avoids mutual interferences, especially when the signals come in the same direction. With this extreme cautious way, I am sure there will be no interferences due to polarization. Thus, I use: Ship - LEO Linear horizontal Linear vertical

LEO -GEO Circular slant 45° Circular slant 135°

GEO-earth station Left-hand circular Right-hand circular

LEO-earth station Horizontal slant 135° Vertical slant 45°

UPLINK DOWNLINK









Arnaud DANNEQUIN

LEO – earth station downlink budget C/N0= (Ptx + Gt.max+ Lf .tx )+(+LFS + Latm-Lrain)+(GRX+ Lpol + k + Ts-Tgalactic) C/N0 =(20+50.6 – 2)+(-177-1.7-141)+(67-3+228.6-26.4-19)=-4 [dBW/Hz]

5.3 Ship to LEO satellite communications Ship-LEO uplink budget C/N0= (Ptx + Gt.max)+(+LFS + Latm-Lrain)+(GRX + Lpol + k + Ts) = (22.4+43)+ (-177-1.7-141) + (50.6-3+228.6-29.6)=-5 [dBW/Hz]

LEO-ship downlink budget C/N0= (Ptx + Gt.max+ Lf .tx )+(+LFS + Latm-Lrain)+(GRX+ Lpol + k + Ts) C/N0 =(20+50.6 – 2)+(-177-1.7-141)+(43-3+228.6-26.4)=2.7 [dBW/Hz]

5.4 LEO to GEO communications LEO-GEO downlink budget C/N0= (Ptx + Gt.max+ Lf .tx )+(+LFS)+(GRX+ Lpol + k + Ts) C/N0 =(27+35 – 2)(-212.7)+(35-3+228.6-29.6) =78.3 [dBW/Hz] Eb/No = C/N 0-10*Log(R b)-Lmargin = 12.3 [dB] -7 -6 BER=1/2*erfc( $(Eb/N0))=3*10 <1.10 LEO-GEO uplink budget C/N0uplink =C/N0downlink =78.3 [dBW/Hz]

RRY100









Arnaud DANNEQUIN Therefore, the Carrier to Noise ratio for uplink between earth station and GEO satellite becomes: -1 C/N0=-59.7 + 94 =34 [dBW/H ] We take a link margin of 3dB for each Carrier to Noise ratio Then, Eb/N0 = C/N0-10*Log(R b)-3=-32 [dB] Unknown BER GEO-earth station downlink: C/N0 =-65.5 +94 = 28.5 [dBW/Hz] Eb/N0 =28.5 -10*Log(Rb)-3=-37.5 [dB] Unknown BER GEO-earth station overall link: Unknown BER Ship-LEO uplink: C/N0=-7.7 + 94 =86.3 [dBW/H] Eb/N0 =86.3 -10*Log(Rb)-3=20.3 [dB] -11 -6 BER=9.34*10 <<10 Ship-LEO downlink: C/N0 =-5 +94 =89 [dBW/H] Eb/N0 =89 -10*Log(Rb)-3=23 [dB] -12 -6 BER=6*10 <<10 Ship-LEO overall link : (86.3/10) -1 (89/10) -1 -9 (C / N0)-1 = (C/N0uplink)-1 + (C/N0downlink)-1=(10 ) +(10 ) =3.5*10 -9 C/N0 =-10*Log(3.5*10 )=84.5 [dBW/Hz] Eb/N0 =84.5 -10*Log(Rb)-3=18.5 [dB] -10 -6 BER=6*10 <<10 LEO – earth station downlink: C/N0 =-4 +94 = 90 [dBW/H] Eb/N0 =90 -10*Log(Rb)-3=24 [dB] -12 -6 BER=2*10 <<10

RRY100











Arnaud DANNEQUIN

RRY100

Conclusion I underestimated the impact and the significant augmentation of the rain fade with the increase of frequency when I first designed my communication system. The latter with these parameters could not provide viable communications in the worst case. By reducing the frequency, my system has become reliable since the quality of the links were better than those required except for the HDTV link largely due to the low gain of the GEO horn antennas. I realized that the main attenuations to take into account are the rain fade and the free space low. Indeed, they affect considerably the satellite communications. On the other hand, this project allowed me to discover and get familiar with satellite communications. With my network mobile background, I have now a complete picture of telecommunications. If I had to make a comparison with network mobile, my personal view is that satellite communications are much more attractive than the first quoted by their complexity and by all parameters to take into consideration. Moreover, it permitted me to gain a method to calculate the quality of a telecommunication link. Even my first designed system was not correct the project allowed me to criticize my results and to review them in order to design a system meeting the specification´s requirements. Some of my calculations perhaps are not fully accurate, but I think the methodology used was done in the right way. Even this project was very challenging, I have succeeded to arrive at the end of it by asking myself a lot of questions and to go on leaving my doubts behind me even I didn’t understand in details modulation and electronic parts. I am personally convinced that both the









Arnaud DANNEQUIN

RRY100

References http://www.sky-brokers.com/home/products/amplifiers/twta/indoor-rackmounted/ku-band/cpi-twta-200w-extendedku-band-12.75-14.50ghz-vzu-6992eb-rackmounted-3ru http://www.ece.gatech.edu/research/labs/bwn/papers/1999/j1.pdf http://www.cobham.com/about-cobham/aerospace-and-security/about-us/satcom/satel http://www.cobham.com/about-cobham/aerospace-and-security /about-us/satcom/satellite-communicati lite-communication-aton-atsea/products-and-services/ku-band-maritime-vsat/ sea/products-and-services/ ku-band-maritime-vsat/sailor-900-vsat/ sailor-900-vsat/sailor-900-vsatsailor-900-vsat-product-sheet.aspx product-sheet.aspx

http://www.cobham.com/media/83787/805-1.pdf http://www.intelsat.com/tools-resources/satellite-basics/satellite-sun-interference/#sthash.SsuhNIIC.dpuf http://www.ieee.li/pdf/viewgraphs/fundamentals_satellite_communication_part_2.pdf https://www.globalstar.com/en/index.php?cid=8400 http://wireless.ictp.trieste.it/handbook/C4.pdf http://www.afcsat.com/microwav.html#ultra

SATCOM Final Report Arnaud Dannequin

Recommend Documents