2018 Twenty Fourth National Conference on Communications (NCC)

System Design Aspects of Ka-Band High

Throughput Satellite (HTS) For Indian Region
Neha Mehra Abhishek Kakkar Subhash Chandra Bera
Space Applications Centre Space Applications Centre Space Applications Centre
Indian Space Research Organization Indian Space Research Organization Indian Space Research Organization
Ahmedabad, India Ahmedabad, India Ahmedabad, India
[email protected] [email protected] [email protected]

Abstract—In the era of advanced communication technology,

there is an ever-increasing demand for providing user-specific
broadband content to a large population. Though ground in-
frastructure is directed towards supporting the high bandwidth
needs in developed regions, the remote and isolated areas remain
devoid of such services. This has led to the evolution of High
Throughput Satellite (HTS) systems to deliver uniform broad-
band connectivity to the developed as well as the unconnected and
remote regions. HTS supports a vast range of user requirements
like video-on-demand, high data rate corporate links, content
distribution/back-haul links from small user terminals. These
satellites are capable of providing large system throughput as
compared to the conventional Fixed Satellite Service (FSS) and
Broadcast Satellite Service (BSS) satellites due to the use of
multi-beam architecture and large system bandwidths. In this
paper, the concept, design and features of a High throughput
satellite system are discussed and the payload configuration for Fig. 1. Trends in High Throughput Satellites.
a geostationary Ka-band HTS over Indian region is presented.
With 1 GHz of total user spectrum in forward as well as return
links, the Indian Ka-band High Throughput Satellite will provide
˜ 80 Gbps of overall throughput with 72 spot beams over India. The concept and implementation of HTS system is a well
investigated subject and lot of literature exists signifying its
I. I NTRODUCTION importance for broadband connectivity to international users
[1] [2] [3] [4] [5]. However, the importance of HTS system for
With the increase in global connectivity and the need to Indian region needs to be recognized. The goal of this paper is
be connected at all the times, bandwidth requirements are to establish the significance of high throughput satellite system
increasing continually. HTS systems have transformed satellite for Indian region and to bring out the payload system design
communications industry in recent years by enabling much aspects for a Ka-band multi-beam satellite system providing
higher capacities at more affordable costs. A high throughput broadband solution to India.
satellite has many times the throughput of a traditional satellite The paper is organized in six sections. Section-II describes
for the same amount of spectrum. The trend in high throughput the significance of HTS for Indian region. Section-III briefly
satellites is towards narrower beams and increased through- explains the concept of high throughput satellite and its com-
puts. These satellites are optimized for high data rate appli- parison with traditional broadcast satellite. Section-IV explains
cations, using multiple spot beams with extensive frequency the system design trade-offs for a multi-beam satellite and
reuse, thus achieving greater capacity than that of conventional section-V presents payload configuration for a Ka-band high
wide-beam satellites used for broadcast applications. throughput satellite system over Indian region. For the con-
There are large numbers of international operational high figured HTS system, RF communication link budget analysis
throughput satellite systems. The first generation HTS satel- and throughput estimation is provided in Section-VI.
lites (like IPStar, Wildblue-1, Spaceway-3) have throughputs
II. S IGNIFICANCE OF HTS F OR I NDIAN R EGION
varying from few Gbps to few tens of Gbps. The second
generation high throughput satellites (like Ka-Sat, Viasat-1, The primary satellite services in India are broadcast services
Jupiter) have throughput ranging from few tens of Gbps to in C-Band and Ku-Band. However, the increasing demand
few hundreds of Gbps. The future high throughput satellites for user-specific broadband content coupled with the limited
are aiming terabit/s capacity. [1] [2] [3] [4] [5] The trends in terrestrial networks in undeveloped regions and tough terrains,
throughput of operational HTS satellites are shown in Fig. 1. has led to the need for a viable broadband solution from space.

978-1-5386-1224-8/18/$31.00 ©2018 IEEE

 2018 Twenty Fourth National Conference on Communications (NCC)

IV. D ESIGN OF H IGH T HROUGHPUT S ATELLITE

A high throughput satellite is a multibeam satellite which
provides high overall system capacity. The HTS system is
primarily bent-pipe architectures and operates in a hub-spoke
mode (user-gateway). It consists of the space segment, which
includes a multibeam communication satellite, and the ground
segment which includes hub stations as well as user terminals.
The fundamental difference in the architecture of an HTS
system is the use of multiple spot beams, rather than single
beam, to cover a desired service area, which brings two-fold
benefit:
• Higher transmit/receive gain
• Intensive frequency re-use
Fig. 2. Traditional Satellite vs HTS.
High throughput satellites have high C/N (carrier-to-noise
ratio) links which can be optimally utilized by employing
advanced modulation and coding techniques [1]. The critical
There is need for a satellite solution to bridge the digital factors and trade-offs for the design of a High Throughput
divide across non-connected and poorly connected areas. Such Satellite system are described in this section.
a satellite would provide uniform quality of service, resolving
last mile issues in rural/ geographically isolated regions. A. Selection of Frequency Band
This will also help to ensure efficient delivery of critical The choice of frequency band for traditional communication
technology-enabled services related to health, education and satellites is C-Band or Ku-Band. Earlier high throughput
growth to a large population. A high throughput satellite satellites were mainly Ku-band multibeam satellites. However,
employing several spot beams over Indian region can prove to the present trend is towards use of Ka-Band multibeam tech-
be a viable solution to meet broadband needs of the country. nology. The use of Ka-Band is advantageous due to following
reasons:
III. C ONCEPT OF H IGH T HROUGHPUT S ATELLITE • As per ITU allocation, the spectrum availability in Ka-
band is more as compared to Ku-band
Traditional Satellites are primarily designed to broadcast
• Ka-band enables narrower beams, higher antenna gains,
same content to many users. These satellites are essentially
improved link budgets and higher throughputs for a given
single-beam satellites providing broadcast or low-data rate
antenna dimension [6]
FSS services over coverage area. Todays consumers want
• Orbital slots are full with Ku-Band satellites; Ka-Band is
individual content, so the challenge is to deliver user-specific,
comparatively less explored and hence orbital slots are
high-bandwidth content to many subscribers over the service
easily available.
area. To support communication through ultra-small aperture
• Ka-band ground terminals produce narrower beams, al-
terminals (USAT), on-board platform should have coverage
lowing closely spaced Ka-band satellites in orbit.
performances to support high data rate services. This calls for
a multi-beam satellite with spot beams covering the desired In future satellites, the use of Q/V-band or higher frequency
region, using high frequency bands such as Ku-band and Ka- bands is envisaged for gaining higher throughputs from space
band [6]. Fig. 2 shows traditional satellite beam vs HTS beams. infrastructure. However, higher frequency bands are associated
with increased atmospheric attenuation and path loss. Hence,
The realization of high throughput satellite involves certain the trade-off involved in selection of frequency band is in-
challenges and critical technologies. Due to large number of creased gain and spectrum availability versus high path-loss
beams, the mass and power requirements of spacecraft are and fading. Ka-band is selected for proposed HTS system due
high. The RF subsystems are miniaturized by using microwave to sufficient spectrum availability, high antenna gain and high
monolithic integrated circuits (MMIC) based technology for throughput. The fading in Ka-band will be managed using fade
active subsystems and advanced materials for passive subsys- mitigation techniques as described in later section.
tems, with stacking and improved packaging to reduce overall
mass and footprint. The communication antennas are light B. Frequency Reuse Scheme
weight with feed clusters to generate multiple beams. Due Frequency reuse defines the number of times a given spec-
to narrow spot beams, the pointing requirements are critical; trum can be re-used over specified service area. Due to mul-
hence RF auto-tracking is used to account for platform stabil- tiple times re-use of same frequency, the interference in this
ity, thermal distortions and residual errors. The system design system is very critical. This interference is characterized by
of high throughput satellites is complicated and involves trade- multi-beam carrier-to-interference (C/I) ratio. The multi-beam
offs at various design stages. C/I ratio is defined as the aggregate carrier-to-interference ratio
 2018 Twenty Fourth National Conference on Communications (NCC)

TABLE I
N UMBER OF B EAMS V S R EFLECTOR S IZE

Options Antenna Dia. Beam Size No. of Beams EOC Gain

(m) (deg.) (dB)
Fig. 3. Four-colour frequency re-use scheme 1 2.2 0.51 56 47.0
2 2.5 0.45 72 48.5
3 2.8 0.40 88 49.5
due to cross-polarization isolation (XPI) of same beam and
beam-to-beam isolation of other co-polarized beams.
The colour is a unique combination of frequency and po-
larization.The common re-use schemes are 3-colour, 4-colour
and 7-colour frequency reuse schemes. It is advantageous to
use less number of colours resulting in higher bandwidth per
beam. However, due to poor beam to beam isolation, the C/I
(carrier-to-interference) ratio decreases, which further limits
the overall throughput [5]. As the number of colours increase,
the interference between beams reduce, but this results in less
bandwidth per beam, reducing the overall system throughput.
(a) 56 beams (b) 72 beams
Hence, there is a trade-off between the overall throughput and
the acceptable amount of interference (C/I). For the proposed
system, four colour re-use is chosen to meet the C/I of
typically 16 dB.
The spectrum allocation to each beam is shown in Fig. 3,
where spatial diversity as well as polarization diversity is used
[1]. Hence, if P1 & P2 are two orthogonal polarizations, F1 &
F2 are two different frequencies, then four colours C1, C2, C3,
C4 are defined as C1:F1P1; C2: F2P1; C3:F1P2; C4: F2P2.
Assuming there are n number of beams over the coverage area,
(c) 88 beams
having C-colour frequency reuse scheme, the frequency re-use
factor (F) can be defined as: Fig. 4. Beam layout over Indian region

F requencyReuseF actor(F ) = n/C (1)

The overall throughput of multibeam system depends on the service area and higher throughput. However, the pointing
aggregate bandwidth (BW) of system. The aggregate band- requirements are stringent for narrow beams. Additionally as
width for forward and return link is defined as: the number of beams increase, the payload hardware increases,
resulting in higher mass and power requirements. Hence, there
AggregateBWF wd = F × T otalSpectrumF wd (2)
is trade-off between the on-board resources in terms of payload
AggregateBWRtn = F × T otalSpectrumRtn (3) mass, power, accommodation as well as achievable pointing
accuracy of the beams and overall system throughput.
The aggregate bandwidth of system is the sum of forward and A trade-off analysis is done for number of beams over
return link bandwidths. Indian region with antenna diameters of 2.2m, 2.5m and 2.8m.
AggregateBWSys = AggregateBW F wd+AggregateBWRtn The beam sizes and typical antenna gain for different reflector
(4) diameters is summarised in Table. I and beam layout is shown
The multibeam satellites achieve higher throughputs as a result in Fig. 4. In proposed HTS system, 72 beam configuration with
of higher aggregate bandwidths 2.5m reflector antenna is chosen to provide reasonable gain
and pointing accuracy requirements. Since beam size is 0.45,
C. Selection of Number of Beams over Service Area the pointing accuracy of the order of 0.04is required. This
The service area for proposed configuration is Indian main- will be achieved using closed loop RF based antenna tracking
land and islands. Total number of beams over service area de- scheme, which accounts the effect of beam de-pointing due to
pends on antenna size and beam-to-beam isolation. The size of platform stability, as well as the antenna surface deformations
antenna is limited by the spacecraft accommodation, whereas due to thermal effects.
beam-to-beam isolation depends on the gain roll-off at the
edge-of-coverage (EOC). These two factors decide the size of D. Rain Fade & Mitigation Techniques
beam. As the antenna size increases, the beam-width decreases The major challenge in using Ka-band for communication
and gain increases, which implies more number of beams over link is its sensitivity to severe atmospheric perturbations which
 2018 Twenty Fourth National Conference on Communications (NCC)

TABLE II
R AIN ATTENUATION & MITIGATION TECHNIQUES

Link Freq. Avbl. Rain Fade* FMT Used

(GHz) (%) (dB)
Forward Link: Hub to user
Uplink 30 99.9 32 ULPC (Hub)
ALC (Transponder)
Downlink 20 99.6 7 Site Diversity (Hub)
ACM
Return Link: User to Hub
Uplink 30 99.6 16 ULPC (User)
Adaptive Coding
Downlink 20 99.9 15 Site Diversity (Hub)
∗ Reference Rain Model is ITU-618.10

include scintillation, cloud attenuation, gaseous absorption

and rain attenuation. Among these factors, rain attenuation is Fig. 5. System Architecture (72 Beams)
most dominant factor and India being a tropical country, fade
mitigation techniques (FMT) are required to compensate high
rain fade. The fade mitigation techniques in HTS system and
the expected gains are:
• Uplink Power control (ULPC) at user terminal (˜10 dB)
& hub station (˜12 dB)
• Site Diversity by using two spatially separated ground
antennas for a hub station [7] (provides ˜12dB gain for
>10 km separation)
• ALC in forward link, which varies the gain of transponder
within ALC dynamic range (typically 20 dB), hence
maintaining the operating point of satellite high power
amplifier (HPA)
• Use of Adaptive Coding and Modulation (DVB-S2X
provides Es/No range varying within ˜26 dB) [8].
Table II shows the typical rain attenuation values over Indian
region and fade mitigation techniques used in various commu-
nication link segments. The frequency in uplink and downlink Fig. 6. Typical Frequency Plan
is 30 GHz and 20 GHz, respectively. The availability is 99.9%
for hub link and 99.6% for user link.
V. S YSTEM A RCHITECTURE Right Handed Circular Polarization (RHCP) are utilized for
The proposed HTS system architecture is shown in Fig. 5. polarization diversity. User spectrum is 250 MHz per beam.
The forward link is defined from hub to user terminal via for employing four colour re-use plan, 500 MHz spectrum is
satellite forward channel; return link is defined from user required in both orthogonal polarizations, resulting in 1 GHz
terminal to hub via satellite return channel. There are 72 of total spectrum. Typical frequency plan is shown in Fig. 6.
user spot beams covering Indian mainland and islands. The The hub spectrum is 1 GHz in both polarizations, resulting in
size of each beam is ˜0.45 deg. There are ten hub stations 2 GHz of total spectrum. This implies that each hub can cater
located within Indian mainland (which includes nine oper- to 8 user beams having 250 MHz band-with per beam. Hence,
ational and one redundant hub). The hub stations perform nine operational hubs are required to cater 72 user beams.
network management, switching and routing. All the hub The frequency re-use factor and aggregate bandwidth can
stations are connected via terrestrial links to enable network be calculated using equations Equation (1) to (4):
and redundancy operations. Frequency Re-use factor = 72/4 = 18
Aggregate BW = 18 GHz (Fwd) + 18 GHz (Rtn) =36 GHz
A. Frequency & Polarization
As per ITU allocations, the available uplink spectrum is 27.0 B. Payload Configuration
- 31.0 GHz, downlink spectrum is 17.7 GHz -21.2 GHz. cicular The payload functional block schematic showing group of
polarization is used since it does not require polarization user beams connected via common hub is shown in Fig. 7.
tracking, resulting in low cost user-terminals. Two orthogonal The complete payload consists of nine such blocks (for nine
polarizations Left Handed Circular Polarization (LHCP) and hubs). The system parameters are described in Table. III.
 2018 Twenty Fourth National Conference on Communications (NCC)

Fig. 7. Functional block schematic for Ka-Band HTS

TABLE III output of DoCons are then passed through input filters (I/P
PAYLOAD S YSTEM PARAMETERS Filters) to provide adjacent channel rejection and remove any
S.No Parameters Units Values unwanted signal before feeding to high power amplifier. The
1 Payload Mass Kg 850 High power amplification is done using 130W Linearized
2
Payload DC Power KW
10.5
channel & Travelling wave tube amplifiers (LCTWTA). The
Requirement output of LCTWTA is de-multiplexed into signal of each
3 Satellite EIRP dBW 69
beam using 2-channel output de-multiplexer. The signal of
4 Satellite G/T dB/k 21
Antenna Reflector m each beam is now passed through high reject filter (HRF) for
5 2.5 rejecting any harmonics produced by high power amplifier and
Dia
No. of Transponders transmitted to corresponding user beams.
6 Forwad Link 36 Return Link Transponders: The user uplink from two
Return Link 18
7 Beam Size deg 0.45
beams having different frequencies are combined after LNA
Modulation Scheme - using 2-channel multiplexer and fed to Down-converter. The
8 Forward Link DVB S2x offset of 500 MHz between DoCon-3 and DoCon-4 ensures
REturn Link DVB RCS2 two staggered frequencies at output of both down-converters.
Two outputs, one from each set of down-converter, are com-
bined using 2-ch multiplexer before feeding to high power
amplifier. The combined signal of 1 GHz is fed to LCTWTA
The 72 spot beams are produced by four parabolic reflector and transmitted to corresponding hub beams.
antennas of diameter 2.5m each. There are 72 transmit/receive As explained above, HPA sharing is used in forward and
feeds with 62 for user beams and 10 feeds catering to both return link transponders; two beams per HPA in forward link
users as well as hub beams. The uplink signals are passed and four beams per HPA in return links. Hence, there are
through Pre-select filters (PSF) to limit the noise bandwidth 36 forward link transponders and 18 return link transponders,
and then pre-amplification is done using Low Noise Amplifiers resulting in total 54 Ka-band transponders. There are two Ka-
(LNA). There are common PSF and LNA for forward and band downlink beacons, one at each polarization to help in
return link transponders. pointing and uplink power control for user terminals and hub
Forward Link Transponders: Each hub uplink consists stations.
of four user beam signals. After low noise amplification,
the hub uplink is passed through 2-channel de-multiplexer VI. L INK B UDGET A NALYSIS
and fed to two different set of down-converters (DoCon-1 The communication link budget analysis for proposed Ka-
& DoCon-2) having LO frequencies in offset of 500 MHz. band high throughput satellite system is described in this
This results in output of DoCon-1 and DoCon-2 at same section. The forward and return links have symmetric band-
frequency, each consisting of two user beam signals. The width i.e. 250 MHz per beam, out of which 225 MHz is
 2018 Twenty Fourth National Conference on Communications (NCC)

usable bandwidth. Forward link uses Digital Video Broadcast TABLE IV

(DVB-S2X) standard with MF-TDMA access scheme. The L INK B UDGET A NALYSIS
DVB-S2X offers optimum spectrum utilization with sharper S.No Parameters Units Values
roll-offs, high efficiency waveforms and increased granularity, Satellite Parameters (Average)
resulting in high throughput within the allocated spectrum 1 EIRPSingleCarrierSaturation dBW 69
G/T dB/k 21
[8]. Return link uses DVB-RCS2 standard with FDMA access
User Terminal (0.8m)
scheme. DVB-RCS2 offers wide range of modcod standards 2 EIRP dBW 48
(as compared to DVB-RCS) with different SNR requirements, G/T dB/k 20
enabling more flexibility to users [9]. Hub Terminal (8.1m)
3 EIRP dBW 75
The maximum data rate achievable at a particular location G/T dB/k 38
within the service area depends on the satellite antenna gain Mod Cod Used
pattern & carrier-to-interference ratio (C/I) and ground ter- 4 Forward Link (DVB-S2x) - 16-APSK 25/36
minal characteristics. Total system throughput is aggregate Return Link (DVB-RCS2 - 16 QAM 3/4
5 Bandwidth per user Beam MHz 225
of data rates achieved at each location within the service
Data Rate
area. For average throughput estimation, the average satellite 6 Forward Link Mbps 50
EIRP (Effective Isotropic Radiated Power) and G/T values are Return Link Mbps 1
considered. The communication link budget analysis for clear Forward Link Calculations
Uplink Path Loss dB 214
sky condition is given in Table IV. The throughput estimation Downlink Path loss dB 210
for 72 beam configuration with 3 GHz of total spectrum is Uplink C/N dB 22
7
given in Table. V. Downlink C/N dB 15
The achievable data rate in forward link is 600 Mbps per C/IM ultibeam dB 17
C/IIM D dB 18
beam, resulting in ˜44 Gbps of total forward link throughput. Total C/(N + I) dB 11
The achievable data rate in return link is 500 Mbps per beam, Return Link Calculations
resulting in ˜36 Gbps total return link throughput. Hence the Uplink Path Loss dB 214
overall system throughput including forward and return links Downlink Path loss dB 210
Uplink C/N dB 18
is ˜80 Gbps. 8
Downlink C/N dB 30
C/IM ultibeam dB 17
VII. C ONCLUSION C/IIM D dB 18
The design, concept and configuration of a High throughput Total C/(N + I) dB 12
satellite is presented and various trade-offs involved in the
system design of a multibeam payload are discussed. A Ka-
TABLE V
band high throughput satellite with 72 spot-beams is config- T HROUGHPUT E STIMATION
ured to provide broadband connectivity over Indian mainland
and islands. The configured satellite is expected to provide S.No Parameters Forward Link Return Link
1 Data Rate /User 50 Mbps 1 Mbps
˜80 Gbps throughput over Indian region. Such a system will
2 No of Carriers / Beam 12 500
bridge the gap between the developed and the under-developed 3 Mod-Cod 16 APSK 25/36 16 QAM 3/4
regions, thus providing uniform broadband service over Indian 4 Data Rate /Beam 600 Mbps 500 Mbps
region. 5 Total Capacity / Beam 44 Gbps 36 Gbps
6 Overall ThroughPut 80 Gbps
ACKNOWLEDGMENT
The authors wish to acknowledge Shri Sumitesh Sarkar,
Group Director, SNSICG, Shri K. S. Parikh, Deputy Director, [4] D. Giggenbach, E. Lutz, J. Poliak, R. Mata-Calvo, and C. Fuchs, “A high-
SNPA, Shri Dipak Kumar Das, Associate Director, Space throughput satellite system for serving whole europe with fast internet
Applications Centre and Shri Tapan Misra, Director, Space service, employing optical feeder links,” in Broadband Coverage in
Germany. 9th ITG Symposium. Proceedings, pp. 1–7, April 2015.
Applications Centre (ISRO) for providing the encouragement [5] H. Fenech, S. Amos, A. Tomatis, and V. Soumpholphakdy, “High
and opportunity to work in this field. throughput satellite systems: An analytical approach,” IEEE Transactions
on Aerospace and Electronic Systems, vol. 51, pp. 192–202, January 2015.
R EFERENCES [6] D. Minoli, High Throughput Satellites (HTS) and KA/KU Spot Beam
Technologies, pp. 95–159. John Wiley & Sons, Inc, 2015.
[1] H. Fenech, A. Tomatis, S. Amos, V. Soumpholphakdy, and D. Serrano- [7] A. Kyrgiazos, B. G. Evans, and P. Thompson, “On the gateway diversity
Velarde, “Future high throughput satellite systems,” in 2012 IEEE First for high throughput broadband satellite systems,” IEEE Transactions on
AESS European Conference on Satellite Telecommunications (ESTEL), Wireless Communications, vol. 13, pp. 5411–5426, Oct 2014.
pp. 1–7, Oct 2012. [8] “Digital video broadcasting (dvb); second generation framing structure,
[2] O. Vidal, G. Verelst, J. Lacan, E. Alberty, J. Radzik, and M. Bousquet, channel coding and modulation systems for broadcasting, interactive
“Next generation high throughput satellite system,” in 2012 IEEE First services, news gathering and other broadband satellite applications; part
AESS European Conference on Satellite Telecommunications (ESTEL), 2: Dvb-s2 extensions (dvb-s2x),” Nov 2015.
pp. 1–7, Oct 2012. [9] “Digital video broadcasting (dvb); second generation dvb; interactive
[3] A. Perez-Carro Rios, R. De Gaudenzi, B. Perrot, and E. Ashford, “New satellite system (dvb-rcs2); part 2: Lower layers for satellite standard,”
generation of european satellite communications systems: Performances, Jan 2012.
analysis and simulation,” 04 2003.