Broadband Non-Geostationary Satellite Communication Systems: Research Challenges and Key Opportunities
BBroadband Non-Geostationary Satellite Communication Systems:Research Challenges and Key Opportunities
Hayder Al-Hraishawi, Symeon Chatzinotas, and Bj¨orn OtterstenInterdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg
Abstract —Besides conventional geostationary (GSO) satellitebroadband communication services, non-geostationary (NGSO)satellites are envisioned to support various new communica-tion use cases from countless industries. These new scenariosbring many unprecedented challenges that will be discussedin this paper alongside with several potential future researchopportunities. NGSO systems are known for various advantages,including their important features of low cost, lower propagationdelay, smaller size, and lower losses in comparison to GSOsatellites. However, there are still many deployment challengesto be tackled to ensure seamless integration not only with GSOsystems but also with terrestrial networks. In this paper, wediscuss several key challenges including satellite constellationand architecture designs, coexistence with GSO systems in termsof spectrum access and regulatory issues, resource managementalgorithms, and NGSO networking requirements. Additionally,the latest progress in provisioning secure communication viaNGSO systems is discussed. Finally, this paper identifies multipleimportant open issues and research directions to inspire furtherstudies towards the next generation of satellite networks.
I. I
NTRODUCTION
Satellites have a distinctive ability of covering wide geo-graphical areas through a minimum amount of infrastructureon the ground, which qualifies them to be an appealing solu-tion to fulfill the growing diversified demand for broadbandservices [1]. Currently, an increasing attention has been paid tothe field of satellite communications from the global telecom-munications market as several network operators start usingsatellites in the backhaul infrastructures for broadband connec-tivity and for the integration with 5G and beyond systems [2].Recently, due to the swift rise of “NewSpace” industries thatare developing small satellites with new low-cost launchers,a large number of satellite operators are planning to launchthousands of non-geostationary (NGSO) satellites to satisfythe burgeoning demand for global high-speed and low-latencyInternet connections [3]. For instance, the emerging NGSOmega constellations such as OneWeb, Telesat, and Starlinkhave a system capacity reaching the terabits-per-second level[4].NGSO satellites on a geocentric orbit include the low earthorbit (LEO), medium earth orbit (MEO) and highly ellipticalorbit (HEO) satellites, which are orbiting constantly at alower altitude than that of geostationary (GSO) satellites, andthus, their link losses are less and the latency due to signalpropagation is lower [5]. These intrinsic features of NGSOsystems besides the high capacities, large footprints, and fastdeployment, offer an interesting set of advantages for the high-speed interactive broadband services that is even competitiveto terrestrial networks [6]. Furthermore, the most newly de- velopments in NGSO systems empower satellites to managenarrow steerable beams covering a relatively broad area, whichfacilities the use of smaller and lower cost equipment at theuser terminals [7], [8]. Specifically, the offered capacities byNGSO satellites can be further increased through utilizing highfrequencies along with employing throughput enhancementtechniques such as spatial diversity, smart gateway, multipleantenna at both user terminals and satellites, and multiple-inputmultiple-output (MIMO) communications [9].These key benefits of NGSO satellites will increase thedensification of their deployments, which might deterioratethe inter-satellite coexistence due to the increased level ofinterference in the shared bands [10]. Therefore, understandingthe interactions between the heterogeneous NGSO satellitesystems is crucial to ensure seamless coexistence. Addition-ally, the emerging NGSO services are expected to operate inthe same frequency bands of the existing GSO systems, whichwill threaten their coexistence in these bands and it will be oneof the challenging issues that will require developing novelsatellite interference mitigation techniques or probably someregulatory interventions [11]. Moreover, due to the highlydynamic feature of NGSO constellations and spatially hetero-geneous traffic demands, the scenario of having heavily loadedsatellite links while others are underutilized will inevitablyoccur frequently, and eventually leads to buffer overflows,higher queuing delays, and significant packet drops at thecongested satellites [12]. Hence, developing load-balancingalgorithms to guarantee a better traffic distribution amongsatellites is indispensable.Notwithstanding the growing interest in NGSO satellites dueto their essential features that can be envisaged for high-speedinteractive broadband services, there are still many brand newchallenges in the NGSO system evolution to be addressed toachieve high quality communications. The objective of this pa-per is to investigate the related forward-looking challenges ofNGSO systems development and integration with highlightingthe applications targeted by NGSO communication systems.By this we aim to discuss the following key implementationaspects: • It is of paramount importance to study communicationsystem design and rearchitecturing directions to integrateNGSO with terrestrial networks to ensure seamless globalcoverage consistent with satellite constellation design. • NGSO systems have also to confront the interferenceissue due to the coexistence with other satellite sys-tems, and thus, developing novel interference coordina-tion/mitigation techniques is crucial. a r X i v : . [ ee ss . SP ] J a n The heterogeneity of NGSO systems alongside with therelative movement between satellites in the lower orbitsmay affect the system performance. Thereby, resourcesmanagement mechanisms are indispensable in such dy-namic propagation environments • The growing number of NGSO satellites increases systemcomplexity and leads to establish space-based networks toimprove coordination and resource utilization. Thus, in-troducing inter-satellite communication links is inevitablein such setups. • The integration of NGSO satellite systems into Internetinfrastructures comes with serious security threats due tothe large constellations that will include hundreds or eventhousands of satellites providing direct connectivity.Beyond this, several prospective future research directions andopportunities for NGSO systems are also provided.II. O
PEN C HALLENGES
Despite the potential advantages offered by NGSO satellitecommunication systems, open essential challenges still needto be addressed.
A. NGSO Satellite Constellation Design
Generally, satellite orbit constellation design is a key factorthat directly affects the performance of the entire satellitesystems. The fundamental constellation parameters includethe type of orbit, altitude of the orbit, number of orbits,number of satellites in each orbit, and satellite phase factorbetween different orbit planes [13]. Several earlier studies haveconsidered systematic constellation patterns of satellites suchas polar constellations and Walker-Delta patterns [14], whichare formulated based on the relative positions of the satellitesin the earth-centered inertial frame (ECI). Additionally, in [15],the concept of flower constellations has been proposed to putall satellites in the same 3D trajectory in the earth-centeredearth-fixed frame (ECEF). However, these design approachesdo not take into consideration the demand characteristics onearth, which makes them inefficient strategies when bearing inmind the non-uniform and uncertain demand over the globe.Thus, a more competent strategy would be a staged flexibledeployment that adapts the system to the demand evolutionand begins covering the regions that have high-anticipateddemands.Another relevant constellation concept that can be appliedto NGSO systems was proposed in [16] to constitute reconfig-urable satellite constellations where satellites can change theirorbital characteristics to adjust global and regional observationperformance. This concept allows establishing flexible con-stellation for different areas of interest. However, introducingreconfigurability feature to the constellation requires a highermaneuvering capability of the satellites and more energy con-sumption and that can be a deterrent factor when multiple suc-cessive reconfigurations are needed over the life cycle. On theother hand, a hybrid constellation design is proposed in [17]to utilize multiple layers and mixed circular-elliptical orbits,and thus, accommodating the asymmetry and heterogeneity of the traffic demand. Nonetheless, the optimization of adaptingthe constellation to growing demand areas is a challengingissue to be addressed in the context of integration an entirehybrid model. Moreover, an integrated framework accountsfor the spatial-temporal traffic distributions and optimizes theexpected life cycle cost over multiple potential scenarios canbe an initial plan to circumvent the NGSO constellation designchallenges.Furthermore, traditional global-based constellation systemsare no longer valid solutions for NGSO systems due to highcost and inflexibility to react to uncertainties resulting frommarket demands and administrative issues. Therefore, regionalcoverage constellations are promising solutions for satelliteoperators as they will be able to tackle the economic and tech-nical issues in a flexible manner [18]. Regional constellationsfocus on the coverage over a certain geographical region byusing a small number of satellites in the system and they canachieve the same or better performance compared to global-coverage constellations. Regional coverage constellations canalso provide sufficient redundancy with deploying multipleNGSO satellites in lieu of a single GSO satellite, and thus,operators can hand off traffic to satellites that avoid beamoverlapping, and therefore avoid interference [19]. However,designing an optimal regional constellation is a complicatedprocess, which requires optimizing the orbital characteristics(e.g., altitude, inclination) while considering asymmetric con-stellation patterns, particularly for complex time-varying andspatially-varying coverage requirements. This topic has notbeen deeply investigated in the literature, and thus, new so-phisticated approaches to design optimal constellation patternsare needed to be developed and tailored to different orbitalcharacteristics and NGSO environments.
B. Coexistence of NGSO and GSO satellites
According to the ITU regulations, the interference inflictedat GSO satellites from NGSO satellite systems shall not de-grade GSO satellites performance and shall not claim protec-tion from GSO systems in the fixed-satellite and broadcasting-satellite services [5]. Specifically, the effective power fluxdensity (EPFD) within the frequency bands that are allocatedto GSO systems and at any point on the earth’s surfacevisible from the GSO satellite orbit shall not exceed the givenpredefined limits in the ITU regulations. Although NGSOsystems have potentials of global coverage and high perfor-mance, many of their regulatory rules were coined nearly twodecades ago based on the proposed technical characteristics ofNGSO satellites at the time. This is very challenging from aspectral coexistence viewpoint, and it will require much moreagile systems. Moreover, the deployment of NGSO satellitesis undergoing a significant densification comparing to existingGSO systems, which is leading to unprecedented inter-satellitecoexistence challenges. The high interference levels will notonly resulting from the enormous number of operating satel-lites but also due to the expected high heterogeneity of theNGSO systems. Therefore, it is imperative to scrutinize thenterference interactions between different GSO and NGSOsystems to ensure consistent hybrid deployment landscape.Despite the several prior works on developing interferencemitigation techniques for satellite systems, the high hetero-geneity and ambiguity about the parameters of the emergingdeployments make the effectiveness of these traditional mit-igation techniques questionable. Moreover, most of the priorworks focus mainly on the inter-system interference betweenGSO and NGSO, while the serious issue of NGSO-NGSOinterference was recently addressed only in [11] and [20].Specifically, the inter-satellite coexistence in the Ku-band isinvestigated in [11] by studying the impact of both NGSO-NGSO and NGSO-GSO co-channel interference on through-put. The impact of NGSO-NGSO co-channel interference onthe achievable throughput for NGSO constellations is studiedin [20]. Band splitting interference mitigation techniques arealso investigated in [20] with considering the Ka and V bands.Accordingly, the highly heterogeneous NGSO constellationproperties and GSO-NGSO interference interactions need tobe thoroughly analyzed for satellite deployments over differentbands and constellations.The concept of mega-constellation brings about spectrumsharing challenges between NGSO and GSO systems. Thesemega-constellation satellites will operate at the same frequen-cies that are currently used by GSO satellites including theKa and Ku bands, which has raised some serious concernsamong GSO satellite operators [2]. Therefore, coordinationand awareness of the operational characteristics about eachcounterpart system is essential in order to achieve a successfulspectrum sharing between different satellites. A database-based operation is foreseen a possible approach can achievesort of coordination between mixed satellite systems [21].Additionally, cognitive satellites with spectrum sensing andawareness can be used in this context along with resourceallocation techniques, i.e., carrier power and bandwidth [22].For instance, designing a cognitive spectrum utilization sce-nario could circumvent the coexistence challenges, whereNGSO satellites can exploit the spectrum allocated to GSOsatellites or terrestrial networks without imposing detrimentalinterference to their concurrent transmissions.Interference analysis of the emerging NGSO constellationsshould take into consideration the effect of the aggregatedinterference due to utilizing a large number of multi-beamsatellites and applying frequency reuse techniques. Fig. 1shows an interference scenario where multiple satellites havingmulti-beam and multi-carrier per beam. Interference to noiseratio (I/N) is an important metric in interference evaluation,and it can be calculated for this scenario as follows
I/N = (cid:80) Si =1 (cid:80) Lj =1 I i,j KB d T d (1)where I i,j represents the downlink single entry interferencelevel at the users on ground from the j -th link of the i -thsatellite, I i,j can be computed as given in [23]. S denotesthe number of interfering satellites and L denotes the numberof interfering links. K accounts for Boltzmann constant, B d GSO SatelliteNGSO coverageGSO coverage NGSO SatellitesInterfering linkFig. 1. Interference scenario involving a GSO and multiple NGSO satellites. represent the downlink bandwidth, and T d is is the receivenoise temperature of the receive antenna. In this context, powercontrol techniques for both the uplink and the downlink canimprove the coexistence performance by mitigating the inter-satellite interference. For example, minimizing the downlinktransmit power P t i of the NGSO satellites can be formulatedas an optimization problem as follows:minimize d ss ∀ S P t i (2a)subject to C i /N i ≥ C /N , ∀ S, (2b) I i,NGSO ≤ I th , ∀ S, (2c)where d ss is the distance corresponding to downlink interfer-ence path with other satellites. C i /N i represents the carrier tonoise ratio of the i -th satellite at the receiver and it shouldbe greater than C /N to keep the link transmitting. More-over, GSO links are protected against the concurrent NGSOtransmissions by maintaining the detrimental co-channel in-terference to an acceptable level that is less than a predefinedinterference threshold ( I th ). C. Resource Management and Orchestration
The high complexity and variety of NGSO satellite archi-tectures and the high-speed mobility with respect to the earth’ssurface inflict multiple resource management challenges thatneed to be carefully addressed. The NGSO satellites aremoving in a higher speed than GSO systems causing morefrequent handovers between satellites during the service [24].Further, the spectrum allocated for the applications that servedby NGSO systems is neither constant nor fully dedicatedduring the service interval. Specifically, the spectrum resourceblocks are allocated based on the available spectrum resources,the speed requirement, and the priority of the service and user.hese enhanced and extended features of NGSO satellites withtheir heterogeneous resources are exacerbating the resourcemanagement challenges. Thus, integrated approaches for spec-trum access strategies that are cognizant of the advantages ofdifferent satellite systems are needed to be flexible and highlyadaptable.Resource management is significantly affected by the em-ployed satellite coverage scheme, where in this regards thereare two popular coverage schemes can be adopted by NGSOsystems: (i) spot beam coverage and (ii) hybrid wide-spotbeam coverage [25]. In a spot beam coverage scheme, eachsatellite provides multiple spot beams to offer coverage overits service area, where their footprint on earth’s surface movesalong with the satellite trajectory. This scheme is simplebut the handovers between beams are more frequent becausethe coverage area of a single spot beam is comparativelysmall. On the other hand, in hybrid wide-spot beam schemeeach satellite provides a wide beam for the whole servicearea and several steering beams for users employing digitalbeamforming techniques. The spot beams are always steeredto the users, and thus, the provided footprint is nearly fixedduring the movement of satellite. In this scheme, handoversoccur between the wide beams only of any two satellites, andthen, the number of handovers substantially decreases due tothe vast coverage of the wide beam.Thereby, the performance of NGSO systems should beoptimized based on its resource utilization, namely, time,power, spectrum, antenna, satellite beams, and orbital planes.When the available NGSO resources are combined with usersand services management over different requirements, severalinteresting resource optimization opportunities arise. Never-theless, formulating and solving such problems become morecomplicated because of the broad resource pool, the interfer-ence issues, and channel state information (CSI) availabilitychallenges. Additionally, resource management schemes tosupport different satellite systems can be designed and tailoredto various objectives, e.g., maximizing the achievable through-put, minimizing the consumed power, reducing latency, andbetter quality of service. In general, development of advancedresource allocation algorithms under different resource avail-ability in the coexistence of NGSO and GSO systems persiststo be an intriguing research direction.
D. NGSO Networking
Currently satellite systems are witnessing a paradigm shiftdue to the rise of NGSO satellites compared to the existedGSO satellites. Specifically, GSO systems are in constant con-tact with ground stations as they control the GSO operations,whereas NGSO systems will need to be built on more au-tonomous and reconfigurable architectures, and the assumptionof persistent contact with ground stations is infeasible in theNGSO case. Thus, NGSO satellites can be deployed as aspace-based network using inter-satellite links, which tendsto be the mainstream of the ongoing NGSO developments.Space-based networks can fulfill the increasing complexityof application requirements. Establishing space-based network architectures is more economically efficient and more suitablefor the heterogeneous integrated satellite communications. Theinter-satellite links in space-based networks eliminate the useof the excessive number of gateways. This architecture is par-ticularly favorable for the areas where acquiring gateway sitesis difficult [10]. Additionally, space-based networks enablecommunication and cooperation between satellites for trafficrouting, throughput maximization, latency minimization, andseamless coverage. These networks have been already usedin some applications, such as navigation, positioning, andweather forecasting satellites, and then, they are anticipatedto play a profound role in NGSO satellite networks and theglobal communication infrastructures in general.The expected better performance of space-based networkswill be achieved at the cost of higher complexity that isnecessary for load balancing between satellite links and forfinding paths with the shortest end-to-end propagation delay,which imposes some restrictions on deploying NGSO net-works for delay-sensitive applications. Given the high-mobilityof NGSO satellites (especially LEO systems) with respect tothe slow moving or fixed user terminals as well as the arbitrarytraffic loads, many concerns are raised about current mobilitymanagement mechanisms and routing algorithms and their effi-ciency to tackle these limitations. Specifically, disregarding theoptimization aspects between traffic distribution and routingstrategies may lead to considerable queuing delays and in-creased number of packet drops at the heavily-loaded satellitelinks [26]. Inter-satellite route planning to satisfy the requiredquality of service levels among users is quite challengingfor the development of NGSO satellite networking, especiallywhen employing a large number of satellites and networktopology is constantly changing. Besides, many parametersare involved for determining the optimal path such as delay,bandwidth, path reliability, link status, traffic load, hop count,etc. Thus, it is essential to design novel routing protocols thatare able to find the optimal route for a data packet to betransmitted between a sender and a receiver while taking intoconsideration the mobility of satellite nodes.
E. Security Challenges
In this subsection, we discuss the security challenges thatneed to be addressed in order to make NGSO satellite systemsmore secure while maintaining seamless interaction with GSOsystems and terrestrial networks. Generally, satellite commu-nications have been relied on terrestrial base stations forprovisioning secured transmissions, which pushed the majorityof security research efforts to focus mainly on the data linksbetween satellites and the terrestrial base stations, i.e., uplinkand downlink [27]. However, the steadily growing deploymentof the space-based wireless network shows that there will bealso a big security risk in the data communication betweensatellites and even the internal structure of satellites. Thesesecurity issues cannot be ignored and they deserve moreattention. Additionally, the complex structure of the space-based wireless network requires various security modeling andnalysis for the space-based NGSO networks in combinationwith certain application scenarios.Proper security mechanisms are essential for NGSO com-munication systems because they are susceptible to securitythreats such as eavesdropping, jamming, and spoofing. Forinstance, any sufficiently well-equipped adversary can sendspurious commands to the satellite and gain full access tosatellites as well as data, enabling them to cause seriousdamage.Another example when satellite systems use strong securitymechanisms for performing message-integrity checks or au-thenticating users, denial-of-service attacks can be conductedby adversaries via sending a large number of spurious mes-sages to the satellite [28]. Thus, satellites under this attack willspend significant computational processing power and time tothe spurious messages, which degrades the quality of servicefor the legitimate users. NGSO satellites can be particularlysusceptible to this kind of attacks because it is a single pointof failure and can be easily overwhelmed if compel to executeexcessive computations.Security of satellite communication is traditionally provi-sioned through cryptographic-based techniques on the upperlayers, which requires high computational complexity [29].Towards this end, on one hand, free space optical (FSO)communication technology is an interesting alternative to RFinter-satellite-links owing to the wide bandwidth and highdata rate that an FSO system can offer, where the opticaltechnologies are foreseen as a key enabler for ultra-securecommunications with the use of, e.g., quantum key distribution[30].On the other hand, as a complementary technique to the tra-ditional cryptographic-based methods, physical layer securityhas been proven to be an effective approach to achieve ever-lasting security without the heavily computational processesof encryption/decryption. Physical-layer security techniquecan be introduced as an added layer of defense into NGSOsatellites, but the studies in this area are still nascent.III. P
OTENTIAL O PPORTUNITIES
In addition to the key challenges that have been raised inthe previous section about NGSO network architectures anddesigns, this section briefly discusses some innovative researchdirections and new opportunities for utilizing NGSO systemsto realize advanced satellite communications for versatileapplications.
A. Space-based Cloud
Far from the common use of satellites as relay devices,the space-based cloud concept has emerged as a promisingand secured paradigm for data storage over NGSO satellites,particularly in the context of big data technologies and appli-cations [31]. The key advantage of space-based data storage isproviding complete immunity from natural disasters occurringon Earth. Furthermore, utilizing NGSO satellites for datastorage can offer more flexibility to some cloud networks thatdesigned to transfer data globally regardless the geographical boundaries and terrestrial obstacles [32]. For instance, mega-corporations and large organizations that are located at dif-ferent global sites can share big data through a space-basedcloud and benefit from the faster transfer rate comparing tothe traditional terrestrial cloud networks, especially for delay-sensitive services. Thereby, NGSO satellites could expandtheir scope of missions for more than only operating as relaydevices for communication networks.
B. IoT via NGSO Satellites
The flexibility and scalability properties of NGSO satellitesmake their employment within the Internet of Things (IoT)ecosystem more appealing to shape novel architectures thatuplift the interoperability among a plethora of applicationsand services [13]. Thus, by exploiting the relatively shorterpropagation distances of NGSO satellite constellations, IoTterminals can be designed to be small-sized, long-life, andlow power consumption, which achieves IoT ideal opera-tion schemes. Moreover, the reduced operating expenditures(OPEX) and capital expenditures (CAPEX) of NGSO satellitescomparing to GSO ones render them into a more feasible wayto deploy efficient IoT services over wide geographical areas[33]. Hence, these exceptional features of NGSO satellitescan unleash the full potentials of IoT, and that will establisha universal network of billions of interconnected devices.However, there are many technical challenges in connectingNGSO satellites to mobile or stationary devices, and this taskparticularly requires a unified network vision on providinghybrid connectivity and prototyping satellite technology tosupport the advancement in machine-to-machine communica-tions and IoT technologies.
C. Caching Over NGSO Satellites
Benefiting from the high-capacity backhaul links and ubiq-uitous coverage, NGSO satellites can help bring content closerto the end users, and thus, these satellite can be consideredas an option for data caching. NGSO satellites also have theability to multi-cast data and quickly update the cached contentover different locations [34]. Additionally, the symbiotic re-lationship between satellite and terrestrial telecommunicationsystems can be exploited to create a hybrid federated contentdelivery network, which will substantially ameliorate userexperience [35]. Therefore, integration of NGSO satellitesinto future Internet with enabling in-network caching makestraffic demands from users for the same content to be easilyaccommodated without multiple transmissions, and thereby,more spectral resources can be saved along with reducingtransmission delay. However, the time-varying network topol-ogy and limited on-board resources in NGSO satellites haveto be taken into account when designing caching placementalgorithms alongside with their fast convergence and lowcomplexity.
D. Terahertz Communications
Terahertz (THz) band communications are anticipated tosupport a wide variety of applications in the upcoming 6Gireless networks [36]. Besides the identified potential indoorTHz use cases and scenarios, there are some foreseen THzcommunication applications within the context of the inte-grated space information networks, e.g., the satellite clusternetworks and inter-satellite backbone networks [37]. Unlikeground THz communications that suffer from short distancetransmission limitations due to the atmosphere attenuation,deploying THz communications in space applications in theatmosphere-free environment circumvents this constraint andachieves high-speed long-distance links between satellites.However, there are still a number of open challenges for THzsatellite communications particularity in terms of semicon-ductor technologies. For example, it is prohibitively difficultto produce high power THz transmitters and current THzreceivers prone to higher noise figures. Thereby, with moreresearch efforts dedicated for developments of high powerTHz transmitters, highly sensitive receivers, and adaptiveantenna arrays, many THz communication opportunities canbe explored within the NGSO satellite deployments.IV. C
ONCLUSIONS
This paper brings up several technical challenges andthought-provoking research directions to integrate NGSOsatellites into the global wireless communication platforms.First, the most foreseen satellite constellation design chal-lenges and rearchitecturing directions are explored to ensureseamless coverage with considering GSO satellites and terres-trial networks. Then, the coexistence challenges in terms ofinterference and spectrum access of NGSO integration withinthe ecosystems of GSO satellites and terrestrial networks arediscussed. The diverse NGSO resources and their managementalgorithms are highlighted as well. Afterwards, the new re-quirements for NGSO networking and the associated securitythreats are deliberated. Furthermore, new application scenariosof the NGSO satellites with novel features and properties arepresented. In short, this paper does not attempt to address everyaspect of NGSO satellites but we are hopeful that it wouldtrigger some more in-depth thinking and enrich the state-of-the-art of NGSO communication systems.A
CKNOWLEDGEMENT
This work is financially supported by the LuxembourgNational Research Fund (FNR) under the project MegaLEO(C20/IS/14767486). R
EFERENCES[1] A. I. Perez-Neira, M. A. Vazquez, M. R. B. Shankar, S. Maleki,and S. Chatzinotas, “Signal processing for high-throughput satellites:Challenges in new interference-limited scenarios,”
IEEE Signal Process.Mag. , vol. 36, no. 4, pp. 112–131, Jul. 2019.[2] G. Giambene, S. Kota, and P. Pillai, “Satellite-5G integration: A networkperspective,”
IEEE Netw. , vol. 32, no. 5, pp. 25–31, 2018.[3] O. Kodheli et al. , “Satellite communications in the new space era: Asurvey and future challenges,”
IEEE Commun. Surveys Tuts. , 2020.[4] Y. Su, Y. Liu, Y. Zhou, J. Yuan, H. Cao, and J. Shi, “Broadband LEOsatellite communications: Architectures and key technologies,”
IEEEWireless Commun. , vol. 26, no. 2, pp. 55–61, 2019. [5] ITU-R, “Simulation methodologies for determining statistics ofshort-term interference between co-frequency, codirectional non-geostationary-satellite orbit fixed-satellite service systems in circularorbits and other non-geostationary fixed-satellite service systems incircular orbits or geostationary-satellite orbit fixed-satellite service net-works,” International Telecommunication Union, Standard, 2003.[6] A. Guidotti, A. Vanelli-Coralli, M. Conti, S. Andrenacci, S. Chatzinotas,N. Maturo, B. Evans, A. Awoseyila, A. Ugolini, T. Foggi, L. Gaudio,N. Alagha, and S. Cioni, “Architectures and key technical challenges for5G systems incorporating satellites,”
IEEE Trans. Veh. Technol. , vol. 68,no. 3, pp. 2624–2639, Mar. 2019.[7] Y. Guan, F. Geng, and J. H. Saleh, “Review of high throughputsatellites: Market disruptions, affordability-throughput map, and the costper bit/second decision tree,”
IEEE Aerosp. Electron. Syst. Mag. , vol. 34,no. 5, pp. 64–80, May 2019.[8] H. Al-Hraishawi, N. Maturo, E. Lagunas, and S. Chatzinotas, “Percep-tive packet scheduling for carrier aggregation in satellite communicationsystems,” in
IEEE Int. Commun. Conf. (ICC) , Jun. 2020.[9] L. You, K. X. Li, J. Wang, X. Gao, X. G. Xia, and B. Ottersten, “MassiveMIMO transmission for LEO satellite communications,”
IEEE J. Sel.Areas Commun. , vol. 38, no. 8, pp. 1851–1865, 2020.[10] N. U. L. Hassan, C. Huang, C. Yuen, A. Ahmad, and Y. Zhang, “Densesmall satellite networks for modern terrestrial communication systems:Benefits, infrastructure, and technologies,”
IEEE Wireless Commun. ,vol. 27, no. 5, pp. 96–103, 2020.[11] C. Braun, A. M. Voicu, L. Simi´c, and P. M¨ah¨onen, “Should we worryabout interference in emerging dense NGSO satellite constellations?”in
IEEE Int. Symposium on Dynamic Spectrum Access Net. (DySPAN) ,2019, pp. 1–10.[12] H. Al-Hraishawi, E. Lagunas, and S. Chatzinotas, “Traffic simulator formultibeam satellite communication systems,” in , 2020, pp. 1–8.[13] Z. Qu, G. Zhang, H. Cao, and J. Xie, “LEO satellite constellation forinternet of things,”
IEEE Access , vol. 5, pp. 18 391–18 401, 2017.[14] J. G. Walker, “Satellite constellations,”
Journal of the British Interplan-etary Society , vol. 37, pp. 559–572, 1984.[15] D. Mortari, M. P. Wilkins, and C. Bruccoleri, “The flower constella-tions,”
The Journal of the Astronautical Sciences , vol. 52, no. 1, pp.107–127, 2004.[16] S. W. Paek, “Reconfigurable satellite constellations for GEO-spatiallyadaptive earth observation missions,” Master’s thesis, MassachusettsInstitute of Technology, 2012.[17] S. Chan, A. T. Samuels, N. B. Shah, J. E. Underwood, and O. L.de Weck, “Optimization of hybrid satellite constellations using multiplelayers and mixed circular-elliptical orbits,” in , Monterey, CA, USA, 2004.[18] H. W. Lee, S. Shimizu, S. Yoshikawa, and K. Ho, “Satellite constel-lation pattern optimization for complex regional coverage,”
Journal ofSpacecraft and Rockets , vol. 57, no. 5, pp. 1309–1327, 2020.[19] H. W. Lee, P. C. Jakob, K. Ho, S. Shimizu, and S. Yoshikawa,“Optimization of satellite constellation deployment strategy consideringuncertain areas of interest,”
Acta Astronautica , vol. 153, pp. 213–228,2018.[20] S. Tonkin and J. P. de Vries, “Newspace spectrum sharing: Assessinginterference risk and mitigations for new satellite constellations,” in
TPRC 46: The 46th Research Conf. on Commun., Inf. and InternetPolicy , 2018.[21] M. H¨oyhty¨a, A. M¨ammel¨a, X. Chen, A. Hulkkonen, J. Janhunen,J. Dunat, and J. Gardey, “Database-assisted spectrum sharing in satellitecommunications: A survey,”
IEEE Access , vol. 5, pp. 25 322–25 341,2017.22] E. Lagunas, S. K. Sharma, S. Maleki, S. Chatzinotas, and B. Ottersten,“Resource allocation for cognitive satellite communications with incum-bent terrestrial networks,”
IEEE Trans. on Cogn. Commun. Netw. , vol. 1,no. 3, pp. 305–317, Sep. 2015.[23] T. Wang, W. Li, and Y. Li, “Co-frequency interference analysis betweenlarge-scale NGSO constellations and GSO systems,” in
Int. Conf. onWireless Commun. and Signal Process. (WCSP) , 2020, pp. 679–684.[24] Y. Li et al. , “An extensible multi-layer architecture model based onLEO-MSS and performance analysis,” in
IEEE 90th Veh. Technol. Conf.(VTC2019-Fall) , 2019, pp. 1–6.[25] E. Lutz, M. Werner, and A. Jahn,
Satellite Systems for Personal andBroadband Communications . Springer Berlin Heidelberg, 2012.[26] R. Radhakrishnan et al. , “Survey of inter-satellite communication forsmall satellite systems: Physical layer to network layer view,”
IEEECommun. Surveys Tuts. , vol. 18, no. 4, pp. 2442–2473, 2016.[27] D. He, X. Li, S. Chan, J. Gao, and M. Guizani, “Security analysis of aspace-based wireless network,”
IEEE Netw. , vol. 33, no. 1, pp. 36–43,2019.[28] A. Roy-Chowdhury, J. S. Baras, M. Hadjitheodosiou, and S. Pa-pademetriou, “Security issues in hybrid networks with a satellite com-ponent,”
IEEE Wireless Commun. , vol. 12, no. 6, pp. 50–61, 2005.[29] Y. Xiao, J. Liu, Y. Shen, X. Jiang, and N. Shiratori, “Secure commu-nication in non-geostationary orbit satellite systems: A physical layersecurity perspective,”
IEEE Access , vol. 7, pp. 3371–3382, 2019.[30] H. Podmore et al. , “Optical terminal for canada’s quantum encryptionand science satellite (QEYSSat),” in
IEEE Int.Conf. on Space Opt. Syst.Appl. (ICSOS) , 2019, pp. 1–5.[31] X. Jia, T. Lv, F. He, and H. Huang, “Collaborative data downloadingby using inter-satellite links in LEO satellite networks,”
IEEE Trans.Wireless Commun. , vol. 16, no. 3, pp. 1523–1532, 2017.[32] H. Huang, S. Guo, and K. Wang, “Envisioned wireless big data storagefor low-earth-orbit satellite-based cloud,”
IEEE Wireless Commun. ,vol. 25, no. 1, pp. 26–31, 2018.[33] M. Bacco, P. Cassar`a, M. Colucci, and A. Gotta, “Modeling reliableM2M/IoT traffic over random access satellite links in non-saturatedconditions,”
IEEE J. Sel. Areas Commun. , vol. 36, no. 5, pp. 1042–1051, May 2018.[34] S. Liu, X. Hu, Y. Wang, G. Cui, and W. Wang, “Distributed cachingbased on matching game in LEO satellite constellation networks,”
IEEECommun. Lett. , vol. 22, no. 2, pp. 300–303, 2018.[35] T. X. Vu, N. Maturo, S. Vuppala, S. Chatzinotas, J. Grotz, and N. Alagha,“Efficient 5G edge caching over satellite,” in , 2018, pp. 1–5.[36] I. F. Akyildiz, A. Kak, and S. Nie, “6G and beyond: The future ofwireless communications systems,”
IEEE Access , vol. 8, pp. 133 995–134 030, 2020.[37] J. Y. Suen, M. T. Fang, S. P. Denny, and P. M. Lubin, “Modeling ofterabit geostationary terahertz satellite links from globally dry locations,”