ACEEE Int. J. on Communication, Vol. 01, No.

03, Dec 2010

Physical Layer Technologies And Challenges In Mobile Satellite Communications.

Sonika Singh1, and R.C.Ramola2
1

Depatment of Electronics & Communication Engineering, Uttarakhand Technical University, Dehradun (U.K), India. gsonika@gmail.com 2 Depatment of Electronics & Communication Engineering, ICFAI University, Dehradun (U.K),India. rramola@gmail.com

Abstract :The central features of the future fourth-generation mobile communication systems are the provisioning of highspeed data transmissions (up to 1 Gb/s) and interactive multimedia services. For effective delivery of these services, the network must satisfy some stringent quality-of-service (QoS) metrics, defined typically in terms of maximum delay and/or minimum throughput performances. Mobile satellite systems will be fully integrated with the future terrestrial cellular systems, playing important roles as back-bones or access satellites, to provide ubiquitous global coverage to diverse users. The challenges for future broadband satellite systems, therefore, lie in the proper deployments of state-ofthe-art satellite technologies to ensure seamless integration of the satellite networks into the cellular systems and its QoS frameworks, while achieving, to the extent possible, efficient use of the precious satellite link resources. This paper presents an overview of the future high-speed satellite mobile communication systems, the technologies deployed or planned for deployments, and the challenges. Index Terms: Mobile satellite Systems, Design issues for MSSs,PHY layer technologies, QoS.

I. INTRODUCTION Satellite networks are an attractive approach for communication services in areas of the world not well served by existing terrestrial infrastructures. There is a vast range of sectors (e.g. land-mobile, aeronautical, maritime, transports, rescue and disaster relief, military, etc.) needing mobile communication services and where the satellite is the only viable option [1]. This is the reason why at present there is a renewed interest and market opportunities for Mobile Satellite Systems (MSSs). Technologies for multispot-beam antennas, low-noise receivers, and on board processing have permitted to achieve the direct access to the satellite for small, portable or even handheld terminals by using S, L, and recently Ku and Ka bands. Satellites can also be equipped with a regenerating payload and intersatellite links, thus respectively permitting to switch traffic ows from different beams of a satellite and traffic forwarding/routing in the sky through satellites. Satellites are on suitable orbits around the earth; on the basis of their altitude, they can be categorized into Geosynchronous Earth Orbit (GEO) and non-GEO. MSSs may suffer from non-Line-of-Sight (non-LoS) propagation conditions due to 28 2010 ACEEE DOI: 01.IJCOM.01.03.2

the presence of obstacles or return link budget restrictions caused by the low power and small antenna size available on portable terminals. In order to address these problems, two similar, but distinct, innovative design approaches can be adopted: (i) hybrid networks and (ii) integrated networks. In the rst case, terrestrial gap llers (repeaters) can be employed to retransmit locally the satellite signal in non-LoS conditions. Moreover, the return link can be supplied by a terrestrial cellular system to simplify the power management of mobile terminals. Finally, the satellite coverage can be extended (e.g. indoor or urban cases) by means of a local wireless system where the base station converts the satellite signal to the wireless one and vice versa. For what concerns the integrated networks, a terrestrial cellular network can be used as an alternative system to connect the mobile user (both forward and return links) with respect to the satellite one. Some examples of integrated networks are analyzed in [2], referring to the Mobile Applications & sErvices based on Satellite & Terrestrial inteRwOrking (MAESTRO) project. In order to dene the terrestrial segment, the European Commission has introduced the concept of Complementary Ground Component (CGC); while, FCC in U.S. has used the term Ancillary Terrestrial Component (ATC). These concepts are quite interchangeable, even if CGC is more related to hybrid networks and ATC to integrated networks. In any case, terrestrial systems could be based on3rd generation (3G), Wireless Fidelity (WiFi, IEEE 802.11 a/b/g), or Worldwide Interoperability for Wireless Microwave Access (WiMAX) technologies. Current Mobile Satellite Systems: The following MSS projects deal with the challenges and the efforts for providing broadband multimedia services to users in land vehicular, aeronautical, and maritime environments: MObile Wideband Global Link sYstem (MOWGLY) [3]; Mobile Broadband Interactive Satellite multimedia Access Technology(MoBISAT) by ETRI [4]; and Broadband Global Area NetworkeXtension (BGAN-X) by the European Space Agency, ESA [5].Moreover, the Satellite-based communication

ACEEE Int. J. on Communication, Vol. 01, No. 03, Dec 2010

systems within IPv6 (SATSIX) project aims, among others, to incorporate the IPv6 protocol inside broadband MSSs [6]. Standards for Mobile Satellite systems: The ve standards that are directly related to MSSs are: Global System for Mobile Communications (GSM) via satellite, SatelliteUniversal Mobile Telecommunications System (S-UMTS),
Digital Video BroadcastingSatellite Version 2(DVBS2) and related return-link standard,

low latency delivery, high capacity, high throughput, and high data rate capabilities. Table 2 shows several mobile satellite systems currently in use which also support highspeed data, Internet, and multimedia applications [7],[8].
Table 1. Satellite Mobile Communication Systems [7],[8].

SatelliteDigital Multimedia Broadcasting (SDMB), and DVBSatellite to Handheld (DVB-SH). Broadband Satellite Architectures and Constellations: Broadband satellite architectures may be based on ATM with sophisticated onboard processing (OBP), onboard switching (OBS), and intersatellite links (ISLs), while others employ simple bent-pipe transponder relays. The system design choices depend on factors including coverage, cost, user service, and traffic demands. Constellations may be LEO, MEO, geostationary earth orbit (GEO), or combinations thereof, dependent on the required coverage and the supported services. Future broadband satellite mobile systems will deploy high numbers of satellites in the nongeostationary constellations, such as MEO and LEO. Though the coverage of GEO satellites is a primary advantage over the LEO systems, the longer delay of GEO however makes them typically less suitable for mainstream 4G applications such as interactive multimedia than the LEO systems. For satellites in LEO, propagation delay is on the order of 10 ms. In MEO the delay is on the order of 80 ms, and in GEO orbits it is 250270 ms. Other delays due to processing and transmissions are on the order of 80100 ms for regional traffics and 140180 ms for international ones. When all delays are considered GEO satellite-based communications may be marginal for quality due to time delays. However, LEO and MEO orbits have their peculiar problems. Due to the low altitude, LEO and MEO satellites move at rapid speeds, causing frequent handovers between the ground terminal and the satellites, which are in view for a relatively short period. The high mobility causes regularchanging network topology and the transmission is subjected to Doppler shifts and small-scale multipath fading. Additionally, LEO and MEO satellites rely on ISLs between neighboring satellites to increase coverage. The main challenge here lies in the proper handling of ISLs so that they do not lead to problems with delay jitter, which can degrade voice and video QoS performances over the satellite systems. Buffering is a good solution known to work well for jitter problems and must be employed. Several satellite mobile systems have been deployed, or are in the process of being deployed, employing specific constellations or mixture of constellations carefully selected to achieve as much as possible, combinations of 29 2010 ACEEE DOI: 01.IJCOM.01.03.2

II. DESIGN ISSUES FOR MSSS NETWORKS. A. Frequency bands and regulations. Frequency bands are assigned at the World Radiocommunication Conferences (WRCs), periodically organized by the International Telecommunication Union radio-communication sector (ITU-R). While xed services use high C and K frequency bands, mobile services are better suited for lower L and S frequency bands that were assigned at the World Administrative Radio Conference (WARC) 92. MSSs have exploited L/Sband technology for a long time: L/S-band systems permit small on-board antennas due to lower signal attenuation and reduced impact of atmospheric effects. However, the need of broadband services and the limited amount of available L/S-band resources (2-30 MHz) have pushed toward the use of Ku and Ka bands for MSSs. ITU-R has assigned Ka band frequency portions to MSSs and Fixed Satellite Systems (FSSs) on a primary basis in all regions (29.930 GHz for earth-to-space link and 20.121.3 GHz for space-to-earth link) and Ku band frequency portions to MSS on a secondary basis in all regions (1414.5 GHz for earth-to-space link and 1012 GHz for space-to-earth link). At present, Ku-based MSSs are available to provide broadband services in many mobile environments, such as trains, boats, planes, and cars. However, Ku-band satellites, as opposed to L/S-band satellites, do not provide a good coverage over seas, because antenna spot-beams footprints are focused on landmasses [9]. In fact, Ku-band satellites are mainly intended for xed users, so that there are not enough Ku/Ka band satellites providing coverage over oceans. Hence, a trade-off has to be achieved between the need of increased bandwidth and coverage issues.

ACEEE Int. J. on Communication, Vol. 01, No. 03, Dec 2010

B.

Mobile terminal antenna. The antenna design is a crucial issue for mobile terminals. An important aspect is the antenna size, the cost, and the adopted technology. Moreover, the antenna system should be reliable and efficient in terms of sensitivity, gain, and interference. It is important to highlight some differences between xed and mobile services: xed terminals use directional antennas, while mobile terminals can also use omni-directional antennas (where ,phasedarray directional antennas with fast tracking algorithms could be adopted instead of omni-directional antennas in order to improve the link budget). Typically, mobile terminals can transmit in all the directions and receive signals from all the directions as well. For this reason, mobile terminals could interfere with other satellite networks. In [17],the study analyzes non-GEO xed and mobile satellite service constellations, providing some suggestions for regulations (in terms of maximum transmitted power and elevation angles) to avoid interference among them. Further considerations on terminal antenna design can be done by taking into account the different application environments: for example, the railway scenario is well served by Ku-band satellites (coverage over landmasses), but the antenna on trains should be small (low-directivity gain), thus generating higher interference levels for adjacent satellites. In aeronautical and maritime scenarios, planes and boats could be at the edge of spot-beam coverage, thus requiring a suitable antenna design. However, big antennas could be used in the case of big boats that have lower design constraints. The antenna size on the mobile terminal determines the characteristics of interference for both uplink and downlink transmissions. Moreover, there are off-axis power ow limitations for uplink transmissions in Ku band (there are only secondary allocations for MSSs). This entails constraints on the Effective Isotropic Radiated Power (EIRP) for the mobile user. In order to mitigate interference, spread-spectrum schemes can be used. Several spread-spectrum techniques can be adopted (e.g. Direct Sequence, DS, Frequency Hopping, FH, and burst repetition). The standardization for the mobile extension of Digital Video BroadcastingSatellite version 2/ Digital Video BroadcastingReturn Channel via Satellite (DVBS2/DVB-RCS) has considered DS spreading for the forward link and burst repetition for the return link (maximum spreading factor of 16 with Single Channel Per Carrier, SCPC) [11]. C.

satellite antennas; dozens of beams for LEO and MEO satellite antennas. The allocated frequency band is divided into some carriers that are distributed among beams in order to avoid interferences among adjacent beams; carriers can be reused in sufficiently far beams. A cluster is a set of beams where all the system carriers are used. Some examples of (average) cluster sizes (i.e. number of beams per cluster) for MSSs are]:12 beams/cluster for Iridium,27 beams/cluster for BGAN, and 21 beams/cluster for Thuraya [14]. GEO systems, such as BGAN and Thuraya, are characterized by higher values of the cluster size: in GEO systems narrower (i.e. higher directivity) beams than in non-GEO ones are needed to irradiate the same area on the earth. Hence, beams are much closer each other in antennas on GEO satellites, thus entailing higher levels of mutual interference and the need for a larger frequency reuse cluster. Here, closer beams means a greater density of satellite antenna beams per unit of solid angle related to the satellite; such density is much higher in GEO cases than in non-GEO ones. On the contrary, the spot-beam footprints (i.e. cells) irradiated on the earth by non-GEO satellites are smaller and closer each other than in the case of GEO satellites. D. Elevation angle. Another important issue for a good quality of the communication is the minimum elevation angle according to which a mobile terminal can see the satellite in an MSS. While the requirements on this angle are not so stringent for FSSs due to the fact that the location and orientation of the user antenna can be optimized (e.g. LoS conditions can be achieved for GEO satellites by selecting appropriate earth station locations), in the MSS scenario (in particular for land-mobile users) a low value of the minimum elevation angle should be avoided , otherwise frequent shadowing and blockage events due to trees, buildings and hills may occur. Increasing the elevation angle, the signal quality improves (reduction of shadowing/blockage effects), but also system costs increase (higher number of satellites in the constellation). The minimum elevation angle requirement entails suitable design constraints for the number of satellites in a constellation and also entails that GEO satellites cannot service Polar Regions. E. Channel models. For the purpose of satellite system analysis, design, and simulation, mathematical models for the land mobile satellite channel are needed. Extensive research works have, therefore, been carried out to develop measurementsbased statistical models [13],[14], that are particularly suitable for Ka- band and higher frequencies. For example, the authors in [15] gave a comprehensive survey of the most accepted statistical models proposed in the scientific literature, considering large-scale and small-fading, single and multiple-state structures, narrowband and wideband channels, and first and second-order statistics. Building upon a thorough characterization of propagation effects, the authors focus on performance analysis of coded and 30

Satellite antenna and frequency reuse. One of the key aspects in realizing MSSs is the use of a high-directivity multi-spot-beam satellite antenna, consisting of a large deployable reector and a feeder system. At present, typical big-antennas on GEO satellites can reach a diameter up to 25 m, a diameter around 2 m can be expected for LEO systems. Spot-beams are needed in order to focus the covered area on the earth with a high antenna gain. Current MSSs exploit satellite antennas with a high number of beams: hundreds of beams for GEO 2010 ACEEE DOI: 01.IJCOM.01.03.2

ACEEE Int. J. on Communication, Vol. 01, No. 03, Dec 2010

uncoded systems based on closed-form expressions, upper bounds, and numerical simulations. Channel Modeling Challenges: One fundamental characteristic of future satellite mobile communication systems is the necessity to be fully integrated into the other terrestrial networks in order to enable global, seamless, and ubiquitous communications. With emerging non-geostationary LEO and MEO satellite systems and high data rate applications, accurate and flexible channel models are needed in order to allow realistic QoS predictions and perform system comparisons under different multiple-access, modulation, coding and diversity schemes. For future satellite mobile systems, a suitable channel model should satisfy the following characteristics: the model should be based on accurate estimation and modeling of propagation statistics, the model should combine very well the effects of weather attenuation process and the multipath fading and shadowing process, and the model should consider the different channel state changes, for example, from a shadowing to a non-shadowing state or vice versa. The choice of channel modeling and estimation should take into account the computational complexity and implementation issues for real time processing. F. Physical (PHY) layer issues

receiving resources in the destination cell, otherwise the related session could be terminated by higher layers. H. Network layer issues With reference to satellites with on-board IP routing capabilities,Mobile IP(MIP) developed by IETF could be used to support handover procedures. Unfortunately, MIP has the problem of high handover latency. NASA and CISCO have carried out many projects to improve MIP for handover procedures in IP-based satellite networks [18]. As an alternative to the above MIP approach, Connexion by Boeing (an in-ight GEO-based Internet connectivity service, not anymore active since 2006) allowed global IP mobility using the Border Gateway Protocol (BGP) [19]. In particular, a Class C IP address block is assigned to a mobile platform (i.e. a plane or a ship, having on-board a data
transceiver/router box and some 802.11 a/b/g wireless access points). These addresses are selectively announced by the nearest terrestrial gateway, for the period the plane/ship passes through the region where the gateway is located (four gateways have been used to cover North America, Europe, and Asian regions). When the plane/ship leaves the region, the gateway stops advertising the IP address block that is advertised by the neighbor gateway. Finally, the IEEE 802.21 Media Independent Handover (MIH) standard could be adopted to manage handovers between IP-based satellite networks and other mobile networks in an integrated system.

An important aspect for MSSs is to use an adaptive air interface with the possible choice among several modulation and coding techniques to adapt to channel variations due to user movement;where adaptation to channel variations implies the use of a feedback channel to inform the transmitter about the most suitable physical layer transmission parameters to guarantee a certain quality at the receiver. Such adoption is viable only for landmobile (low speed) users and becomes critical for higher frequency bandsThe signal blockage effects can cause a demodulator synchronization loss with a period of unavailability during the resynchronization process. Different solutions may be used to face this non-LoS problem: for example, gap llers (in the presence of extended or permanent obstacles), space diversity (e.g. using two receiving antennas that are distant more than the length of obstacles), and time diversity (e.g. using a time interleaver for spreading the errors occurring during a persistent fading event). G. Medium Access Control (MAC) layer issues According to [18], many handover scenarios can be considered: in a non-GEO case, user mobility is dominated by the satellite constellation mobility; while in a GEO case, mobility is present only for users accessing the service from planes, trains, and ships. The resource assignment at the MAC layer (layer 2) has to provide adequate priorities for handover management: handed-over traffic typically suffers from extra switching delays (and, in some cases, rerouting delays when gateway changes are involved) and, hence, it needs an adequate layer 2 prioritization in 31 2010 ACEEE DOI: 01.IJCOM.01.03.2

III. ISSUES INVOLVED IN QOS SUPPORT: The main issues involved in QoS provisioning is the fact that different traffic types have different QoS
requirements, which results in different service levels. Pocketsize voice traffic is characterized as relatively low bandwidth (typically 8 Kb/s), but requires very low latency delivery to ensure high-quality audio at the destination. Such traffic for example, is tagged high priority to protect its service quality. Video traffic, on the other hand, generally has higher bandwidth (128 to 384 Kb/s or more), but still similarly require low latency for high-quality video images at the destination. Data traffics like file transfer, e-mail messages, etc., can generally be allowed to suffer latency through the network without appreciable QoS deterioration. While an e-mail message is typically low bandwidth, file transfer takes significantly high bandwidth. The goal of resource management for QoS is to share properly and efficiently access to the available resources among these different traffic types with the aim of keeping their required quality. Large packets delivered from a high-bandwidth delay tolerant data service like file transfer, for example, may cause qualitydegrading delay to latency intolerant application such as voice. If a 1500-B packet delivered as part of a file transfer over a 64 Kb/s link will take 187 ms to be transmitted, voice and video packets in queue behind this data packet must keep waiting for this time interval. As a result, voice cuts will be heard for the voice traffic, while jitter may be observed in the video images. Effective resource sharing mechanisms, therefore, play important roles in QoS provisioning.

ACEEE Int. J. on Communication, Vol. 01, No. 03, Dec 2010

IV. CONCLUSIONS: Currently, there is a renewed R&D interest for MSSs due their capabilities to provide services anytime and anywhere. This paper surveyed the current mobile satellite networks and services with respect to design issues, physical layer technologies and challenges, recent standardization advances (eg. Mobile extension for DVBS2/RCS,DVB-SH) and some operational systems (eg Globalstar, Inmarsat BGAN, Iridium and Thuraya.). The paper discusses the available frequency bands for the current MSSs, the design characteristics of mobile terminal antenna and satellite antenna. The paper discussed the minimum elevation angle requirement for land mobile users and channel models to characterize a mobile satellite system. The last part of the paper discusses the important design issues for physical layer, medium access control layer , network layer and QoS requirements. Satellite services in previous-generation systems were limited to low bitrate applications. In the 4G system, the trend is toward global information networks offering flexible multimedia information services to users on demand, anywhere, anytime. Satellite-based mobile systems will be used in this regard in a complementary mode to the terrestrial system to meet user demands better. Broadband satellite links will also be used as the backbone in the global network, providing ubiquitous multimedia and high-speed data applications. ACKNOWLEDGEMENT: The work presented in this paper has been carried out within the framework of the European Sat NEx II(Contract No. IST-027393), Network of Excellence on Next Generation Satellite Communications. The authors would like to thank Professor Dharmendra Singh(I.I.T Roorkee, India) for his continuous support and valuable suggestions. REFERENCES
[1]. Kota SL, Pahlavan K, Leppanen P.Broadband Satellite Communications for Internet Access. Kluwer Academic Publishers: Hingham, MA, 2004. [2] Andrikopoulos I, Gallet T, Widmer H, Dubois T, Larzul P-Y, Pouliakis A. Satellite digital multimedia broadcastingexperimentation and validation. Proceedings of the Advanced Satellite Mobile Systems 2006 (ASMS 2006), Herrsching am Ammersee, Germany, May 2006. [3] MOWGLY project Web site available on-line at the URL: http://www.mowgly.org/. [4]. ETRI Web site available on-line at the URL: http://www.etri.re.kr/eng. [5]. Richharia M, Trachtman E. Inmarsats broadband mobile communication system. Workshop Entitled Advances in Satellite

Communications: New Services and Systems, Co-located with Globecom 05, St. Louis, MO, December 2,2005. [6]. Martinez RM, de Domingo MC, Guerra Exposito JA. SATSIX PROJECT: a rst approach to IPv6 over Satellite Networks. Proceedings of the 16th IST Mobile and Wireless Communications Summit, 2007, July 15, 2007; 14. [7] J. Farserotu and R. Prasad, A survey of future broadband multimedia satellite systems, issues and trends, IEEE Commun. Mag.,vol. 38, pp. 128133, June 2000. [8] A. Jamalipour, Broad-band satellite networksThe global IT bridge, Proc. IEEE, vol. 89, pp. 88104, Jan. 2001. [9]. Arcidiacono A, Finocchiaro D, Grazzini S. Broadband mobile satellite services: the Ku-band volution.Proceedings of the 2006 Tyrrhenian International Workshop on Digital Communications(TIWDC06), Island of Ponza, Italy, September 58, 2006. [10]. Henri Y. Non-GSO MSS/FSS constellations and the international regulations, Regional Radio communication Seminar, Mexico City, Mexico, September 2428, 2001. [11].DVB, Interaction channel for satellite distribution systems, BlueBook A054r4.1, January 2009, available on-line at the URL: http://www.dvb.org/technology/standards. [12]. ASMS-Task Force Technical Group, Overview of existing standards and architectures, Internal Report, May 2001, available on-line at the URL: ftp://ftp.cordis.europa.eu/pub/ist/docs/ka4/asms_01_t05_0.doc. [13] C. Loo and J. Butterworth, Land mobile satellite channel measurements and modeling, Proc. IEEE, vol. 86, pp. 1442 1463, July 1998. 14] D. Lutz, M. Cygan, F. Dippold, M. Dolainsky, and W. Papke, The land mobile satellite communications channelrecording,statistics, and channel model, IEEE Trans. Vehicular Technology., vol. 40,pp. 375386, May 1991. [15 ] H. Wakana, Statistical models for land mobile and fixed satellite communications at Ka band, in Proc. IEEE Vehicular Technology Conf.,1996, pp. 10231027. [16] Y. Karasawa, K. Kimura, and K. Minanmisono, Analysis of availability improvement in LMSS by means of satellite diversity based on three state propagation channel model, IEEE Trans. Vehicular Technology , vol. 46, pp. 10471056, Nov. 1997. [17] H. Wakana, Propagation model for simulating shadowing and multipath fading in land-mobile satellite channel, IEE Electron. Letter .,vol. 33, pp. 19251926, Nov. 1997. [18] M. Rice and B. Humpherys, Statistical models for the ACTS K-band land mobile satellite channel, in Proc. IEEE Vehicular Technology Conf., vol. 1, 1997, pp. 4650. [19] C. Caini and G. E. Corazza, Satellite diversity in mobile satellite CDMA systems, IEEE J. Select. Areas Commun., vol. 19, pp. 13241333, July 2001. [20]. Atiquzzaman M, Chowdhury PK, Ivancic W. SIGMA for seamless handover in space. Sixth Annual NASA Earth Science Technology Conference, College Park, MD, June 2729, 2006. [21]. Dul AL. Global IP network mobility using Border Gateway Protocol (BGP). White Paper, available on-line at http://www.quark.net/docs/Global_IP_Network_Mobility_using_ BGP.pdf.

32 2010 ACEEE DOI: 01.IJCOM.01.03.2