LEO Satellites in 5G and Beyond Networks: A Review From A Standardization Perspective
LEO Satellites in 5G and Beyond Networks: A Review From A Standardization Perspective
LEO Satellites in 5G and Beyond Networks: A Review From A Standardization Perspective
net/publication/359492529
CITATIONS READS
39 1,015
5 authors, including:
All content following this page was uploaded by Gunes Karabulut Kurt on 28 June 2022.
Abstract—Low Earth Orbit (LEO) Satellite Network (SatNet) and airplanes—are two of the main challenges for future
with their mega-constellations are expected to play a key role telecommunication systems.
in providing ubiquitous Internet and communications services Satellite communication networks utilize spaceborne plat-
arXiv:2110.08654v1 [cs.NI] 16 Oct 2021
A. Motivation and contributions potential solutions. Section VI sheds the light on the manage-
ment and orchestration aspects of 5G networks with integrated
In the literature, there are several reviews and surveys on
satellite access components. The activities of the non-3GPP
satellite systems. Table I provides a comparative overview
standardization organizations are discussed in Section VII.
of the existing surveys that discuss standardization efforts in
Section VIII highlights important standardization directions
the area of SatNets. The most common topic among existing
that are required for the full integration of satellite networks
surveys is the different 3GPP architectures for 5G with satellite
with 6G networks. Section IX draws the essential conclusions.
access. Nevertheless, the discussion on SatNet related 3GPP
work therein stays at a high level as exiting surveys objective is
to give a broad view of the SatNet state-of-the-art. In contrast, II. S ATELLITE ACCESS N ETWORK E LEMENTS AND
this article aims to offer a dedicated and comprehensive review C HARACTERISTICS IN 3GPP S TANDARDIZATION
of the 3GPP standardization work in the area of SatNets. The
Due to the wide service coverage of SatNets and their
scope of this review spans across multiple areas, from radio
reduced vulnerability to physical attacks and natural disasters,
access network to services and system aspects. Regarding the
the 3GPP sought to define the expected role of SatNets in 5G+
contributions, this work:
through the following points in TR 38.811 [10]:
• Provides a comprehensive survey of the 3GPP standard-
• Provide 5G service in unserved areas that cannot be
ization activities in the area of satellite networks and
covered by terrestrial 5G networks (e.g., isolated and
communication from Release 14 to Release 18.
remote areas, on aircrafts and ships) and underserved
• Highlights the 3GPP use cases for satellite access in 5G
areas (e.g., suburban and rural areas).
and their applications.
• Upgrade the performance of limited terrestrial networks
• Discusses the required adaptation of NR for SatNets from
in a cost-effective manner.
the 3GPP perspective.
• Support the reliability of 5G service by providing service
• Summarizes the potential requirements for the manage-
continuity for M2M/IoT devices or for passengers on-
ment and orchestration of integrated satellite components
board moving platforms and ensuring service availability
in a 5G network.
anywhere, especially for critical communications and
• Presents an overview of standardization efforts from
railway, maritime, and aeronautical communications.
organizations other than the 3GPP.
• Enable 5G network scalability by providing efficient mul-
• Discusses future directions to be taken in standardization
ticast and broadcast resources for data delivery towards
efforts for 6G satellite communication networks.
the network edges or even user terminal.
As mentioned in 3GPP TR 38.811, satellite access networks
B. Paper organization consist of the following elements [10]:
The remainder of this paper is organized as follows. Sec- • User Equipment (UE) or a specific terminal to the satellite
tion II gives a brief description of the satellite access network system in case the satellite doesn’t serve UEs directly.
elements and highlights the characteristics of satellite networks • A service link which is the radio link between the UE
from the perspective of 3GPP. Section III gives an overview and the space platform.
of 3GPP standardization activities with respect to satellite • A space platform carrying a payload which may have one
networks. Section III-B describes the satellite access networks of these two configurations:
use cases in the context of 5G. The architectures of integrated – A bent-pipe payload that performs radio frequency
satellite access networks and 5G networks are presented in filtering, frequency conversion, and amplification.
Section IV. Section V discusses the adaptation of the New – A regenerative payload offering radio frequency fil-
Radio for satellite networks. In particular, subsection V-A tering, frequency conversion, and amplification as
specifies the constraints associated with satellite networks and well as demodulation and decoding, switch and/or
subsection V-B highlights the NR impacted areas and their routing, coding and/or modulation. It is equivalent
3
to how a base station functions (e.g., gNB) on-board efficiently through satellite systems, such as broadcasting ser-
a satellite. vices and delay-tolerant services. In 3GPP TS 22.261 Release
• Inter-satellite links (ISLs) in case of a regenerative pay- 15, “Service requirements for next generation new services
load and a constellation of satellites. An ISL may operate and markets,” the first analyses that described the significant
in RF frequency or optical bands. role that satellites could play in 5G systems were provided.
• Gateways that connect the satellite access network to the However, the focus of TS 22.261 was on satellites in NR for
core network. industrial and mission critical services.
• Feeder links which refer to the radio links between the The increasing interest in integrating satellites with 5G led
gateways and the space platform. to the definition of NTNs by 3GPP. Although NTNs include
In 3GPP TR 22.822 [11] the orbits that are used for other aerial systems (e.g., HAPSs), the 3GPP community
communication satellites are calssified into the following four considers satellites as the main case and other aerial systems
types: as a special case of satellites. Release 16 is frozen and Release
17 is still open with an expected deadline on December
• GEO satellites, located precisely along the plane of the
2021. Release 17 is working on NTNs for 5G systems, which
Equator at an altitude of 35,786 km. GEOs orbit at
adopt satellites to support underserved areas (e.g., isolated and
the same rate as the earth’s rotation. Therefore, a GEO
remote areas, onboard aircrafts and vessels).
satellite can provide continuous coverage.
• Non-Geostationary Orbiting (NGSO) satellites which do
not stand still with respect to the earth. To provide A. 3GPP standardization activities for satellite access net-
service continuity over time, a number of satellites (a works
constellation) is required to meet this requirement; the Through Releases 15, 16, 17, and 18, 3GPP launched
lower the altitude, the higher the number of satellites. several standardization activities to support the integration of
Different types of NGSO satellites are listed below: SatNets and 5G terrestrial networks. The following points
– LEO satellites, with altitudes ranging from 500 km give an overview of the satellite-related 3GPP standardization
to 2,000 km, and with inclination angles of the activities:
orbital plane ranging from 0 up to 180 degrees • Release 15: In 2017, two SIs were initiated: (1) 3GPP
(prograde and retrograde orbits). These constellations TR 38.811 “Study on NR to support Non-Terrestrial
are situated above the International Space Station and Networks” under the RAN TSG; and (2) 3GPP TR
debris, and below the first Van Allen belt. 22.822 “Study on using Satellite Access in 5G” under
– MEO satellites, with altitudes ranging from 8,000 the SA TSG. The first SI aimed to define the NTN
to 20,000 km. The inclination angles of the orbital deployment scenarios and their system parameters as well
plane range from 0 up to 180 degrees (prograde and as to identify the required NR adaptation to accommodate
retrograde orbits). These constellations are situated NTNs. Also preliminary solutions were introduced to
above the Van Allen belts. address the impacted areas of NR. Although the second
– Highly Eccentric Orbiting (HEO) satellites, with a SI was initiated in 2017, it was moved to Release 16.
range of operational altitudes between 7,000 km • Release 16: The SA TSG had four activities: (1) an SI on
and more than 45,000 km. The inclination angle is “Study on using satellite access in 5G” [11]; (2) an SI on
selected so as to compensate, completely or partially, “Study on architecture aspects for using satellite access
for the relative motion of the earth with respect in 5G” [12]; (3) a WI on “Integration of Satellite Access
to the orbital plane, allowing the satellite to cover in 5G” (WI#800010-5GSAT, Release 16); and (4) an SI
successively different parts of northern land masses on “Study on management and orchestration aspects with
(e.g. Western Europe, North America, and Northern integrated satellite components in a 5G network” [13].
Asia). The first and second SIs introduced a number of use cases
for the provision of services in the integrated 5G and
III. OVERVIEW OF S ATELLITE ACCESS N ETWORK 3GPP satellite-based access components. This led to identifying
S TANDARDIZATION AND U SE C ASES corresponding modified or new requirements related to
From the early stage of 5G standardization activities, 3GPP connectivity, roaming, QoS, UE, security, and regulatory.
considered the integration of satellite systems to be a valuable Finally, the most critical issues (and possible solutions)
asset to complement and integrate terrestrial NR networks. related to the management and orchestration of 5G with
Therefore, several Study Items (SIs) and Work Items (WIs) integrated satellite components were addressed in [13].
were initiated within the Technical Specification Group (TSG) With respect to the NR 3GPP activities the SI “Solutions
of the Radio Access Network (RAN), the Service and System for NR to support Non-Terrestrial Networks (NTN)” [14]
Aspects (SA), and the Core Network and Terminals (CT). was completed at the end of 2019. This SI completed
As an extension to terrestrial networks, satellites were first the work that was initiated in Release 15 [10]. A set
mentioned in a deployment scenario of 5G in 3GPP TR of required adaptations to enable NR technologies and
38.913 Release 14. This was to provide 5G communication operations in satellite networks were addressed, covering
services for areas where terrestrial coverage was not available several issues in RAN1 (Physical layer), RAN2 (Layer 2
and also to support services that could be accessed more and 3), and RAN3 (Interfaces).
4
• Release 17: Work on the SI entitled “Study on architec- – Sidelink enhancements (excluding positioning)
ture aspects for using satellite access in 5G” [12] contin- – RedCap evolution (excluding positioning)
ued and was last updated in March 2021. By the end of – NTN evolution, including both NR and IoT aspects
2019, two WIs for NTNs had been initiated: (i) “Solutions – Evolution for broadcast and multicast services
for NR to support NTN” [14], under RAN activities; – Expanded and improved positioning
and (ii) “Integration of satellite components in the 5G – Evolution of duplex operation
architecture” [15], under SA. For the former, the activities – Artificial Intelligence (AI)/Machine Learning (ML)
are in the final stage and it was last updated on June 2021. – Network energy savings
However, the objective is the following points: (i) consol- – Additional RAN1/2/3 candidate topics:
idation of the impacts on the physical layer and definition ∗ Set 1: UE power savings, enhancing and extending
of potential solutions; (ii) evaluating the performance the support beyond 52.6GHz, Carrier Aggregation
of NR in selected deployment scenarios (LEO based (CA)/Dual-Connectivity (DC) enhancements (e.g.,
satellite access, GEO based satellite access) through link Multi-Radio/Multi-Connectivity (MR-MC), etc.),
level (radio link) and system level (cell) simulations; and Flexible spectrum integration, Reconfigurable In-
(iii) identification of the potential requirements for the telligent Surfaces (RIS).
upper layers based on the considered architectures. The ∗ Set 2: UAV, Industrial Internet of Things
latter WI goal was to extend the analysis provided in (IIoT)/Ultra-Reliable Low-Latency Communica-
[12] through the following: (i) identification of impacted tion (URLLC), <5MHz in dedicated spectrum,
areas in NR systems due to the integration of satellite other IoT enhancements and types, HAPSs, Net-
components in 5G; (ii) analysis of the issues related to work coding.
the interaction between the core network and the RAN; ∗ Set 3: Inter-gNB coordination, network slicing
and (iii) identification of solutions for the two highlighted enhancements, multiple universal subscriber
use cases (terrestrial and satellite network roaming and identity modules), UE aggregation, security
5G fixed backhaul). Several areas were considered, such enhancements, Self-Organizing Networks
as network discovery and selection and network slicing. (SONs)/Minimization of Drive Test (MDT).
For Release 17, the freeze of the RAN1 physical layer – Potential RAN4 enhancements.
specifications is scheduled to be in December 2021. This
The importance of the NTN evolution has been very visible
is to be followed by the Stage 3 freeze (RAN2, RAN3
in RAN meetings, including RAN#93-e. In addition, satellite
and RAN4) by March 2022 and the ASN.1 freeze and
access might be discussed under other topics, such as mobility
the performance specifications completion is planned for
management and evolution for broadcast and multicast ser-
September 2022 based on the timeline agreed back in
vices. A summary of the 3GPP satellite related standardization
December 2019.
activities is presented in Table II.
• Release 18: The June 28 – July 2, 2021 Workshop
on Release 18, which is the start of 5G-Advanced, is
to specify the topics of Release 18 with submissions B. Use cases for satellite access in 5G
divided into three areas (preliminary agenda): Enhanced In March 2017, as part of Release 14, the 3GPP initiated
Mobile Broadband (eMBB) driven work, Non-eMBB an SI to analyze the feasibility of satellite integration into
driven functionality and cross-functionality for both. This 5G network (3GPP TR 22.822 [11]). The initial goal was to
3GPP workshop on the radio specific content of Release bring together satellite operators and other companies to create
18 reviewed over 500 presentations by companies and aligned contributions in the support of satellites in the 5G
partner organizations, to identify topics for the immediate standardization. The two SIs 3GPP TR 38.811 and 3GPP TR
and longer-term commercial needs. The Release 18 Pack- 38.821 had been already completed. They studied the role of
age Approval is expected at the December 2021 Plenary satellites in the 5G ecosystem. In addition, the challenges of
(RAN#94). The time duration for the release in RAN a co-existing satellite-terrestrial network had been analyzed
is tentatively set at 18 months. The detailed discussions taking into account different architectural options.
on how to consolidate topics into WIs and SIs will not On the basis of satellite networks characteristics introduced
begin before RAN#93-e meeting in September 2021. This in Section II above, three main use case categories are defined,
meeting will see progress on ‘high-level descriptions’ of namely service continuity, service ubiquity, and service scala-
the objectives for each topic. The list of topics include bility, as shown in Figure 1. Table III summarizes the satellite
the following [16]: access use cases in [10] and [11] and their applications. To
realize these use cases, some requirements need to be fulfilled
– Evolution for downlink Multiple-Input Multiple- by 5G+ systems:
Output (MIMO) • The 5G system shall support service continuity between
– Uplink enhancements land-based 5G access and satellite-based access networks
– Mobility enhancements owned by the same operator or by an agreement between
– Additional topological improvements (Integrated Ac- operators with guaranteed QoS while switching to or from
cess Backhaul (IAB) and smart repeaters) terrestrial to satellites.
– Enhancements for Extended Reality (XR) • Provide the optimum network selection.
5
Fig. 1: Satellite access use cases in 5G. Service continuity use cases provide continuous access to services granted by the 5G
system, while users move between terrestrial and satellite networks. Use cases considering fleets of such UE (whether locally
grouped or dispersed) are also included in this category. Service ubiquity use cases serve potential users wishing to access
5G services in “unserved” or “underserved” areas by terrestrial networks, which will be possible through 5G satellite access
network service. Service scalability use cases utilize the distinguished capability of satellites in multicasting or broadcasting a
similar content over a large area, and potentially directly to user equipment. Similarly, a satellite network can also contribute
to off-loading traffic from terrestrial networks during busy hours by multicasting or broadcasting non-time-sensitive data in
non-busy hours.
• Support mMTC and Nb-IoT services. satellite access with RTT of 30-50 ms [17]. These RTT
• The 5G system shall support the use of satellite links values include the delays of processing on both ground
between the radio access network and core network and and orbit as well as the variable propagation delays. In
within the core network by enhancing the 3GPP system satellite communication, propagation delay varies due to
to handle the latencies introduced by satellite backhaul. changes in satellite and user positions, which lead to
• Reciprocal cooperation is required between mobile oper- different slant ranges.
ators and satellite operators to ensure good service areas • UE with satellite access shall have the capability to accept
available for customers. or reject connections with the satellite on the basis of the
• A 5G system supporting 5G satellite access and 5G quality of class of indicators that are supported, and on
terrestrial access shall be able to optimally distribute user the basis of the available accesses.
traffic over both types of access. • A 5G system with multiple access shall be able to
• A 5G system providing service with satellite access shall select the combination of access technologies to serve
be able to support QoS indicators adapted to GEO-based UE on the basis of the targeted priority, pre-emption, QoS
satellite access with RTT of 600-800 ms, MEO-based parameters and access technology availability.
satellite access with RTT of 125-250 ms, and LEO-based
6
Data Network
NR-Uu NR-Uu N1/2/3 N6
In TR 38.811 [10], the 3GPP community introduced two
(a) NG-RAN architecture with transparent satellite
types of satellite access networks:
Satellite NG Radio access network
• A broadband access network serving Very Small Aperture UE
gNB-DU
Remote radio
unit gNB-
5GCN
CU
Terminals (VSATs) that can be mounted on a moving
F1 over Data Network
platform (e.g. bus, train, vessel, aircraft). In this context, NR-Uu Satellite Radio
Interface
N1/2/3 N6
broadband refers to at least 50 Mbps data rate and even (b) NG-RAN with a regenerative satellite based on gNB-DU
up to several hundreds Mbps for downlink. The service Satellite NG Radio access network
UE Remote radio
links operate in frequency bands allocated to satellite and gNB
unit 5GCN
aerial services (fixed, mobile) above 6 GHz. N1/2/3 over Data Network
NR-Uu Satellite Radio N1/2/3 N6
• Narrow- or wide-band access network serving terminals Interface
equipped with omni- or semi-directional antenna (e.g., (c) NG-RAN with a regenerative satellite based on gNB processed payload
Satellite NG Radio access network
handheld terminal). In this context, narrow-band refers to UE Radio frequency Remote radio
bent pipe unit gNB 5GCN
less than a 1 or 2 Mbps data for downlink. The service
links operate typically in frequency bands allocated to NR-Uu NR-Uu N1/2/3 N6
Data Network
3
2/
NR-Uu
1/
N
In terms of architecture, Figure 2 shows a comparison
between a typical satellite network physical architecture and (d) Multi connectivity involving transparent satellite-based NG-RAN and cellular NG-RAN
Satellite NG Radio access network
a 3GPP NG-RAN architecture. To integrate satellite access UE Radio frequency Remote radio
bent pipe unit
networks in 5G, 3GPP TR 38.821 introduced the following gNB 5GCN
/3
The satellite payload implements frequency conversion bent pipe unit
/2
-U
gNB
N1
and a radio frequency amplifier in both uplink and u
NR-Uu
downlink directions. Several transparent satellites may be
(e) Multi connectivity between two transparent satellite-based NG-RAN
connected to the same gNB on the ground. Satellite NG Radio access network
UE 5GCN
• Regenerative satellite-based NG-RAN architectures: gNB
Remote radio
unit
Data Network
The satellite payload implements regeneration of the NR-Uu N1/2/3 N6
N1/2/3 over
signals received from earth. The satellite payload also Satellite Radio
Interface
N
3
R
/2/
-U
N1
u
Remote radio
interface or an optical interface that may be 3GPP or non- gNB
unit
Typical satellite network physical architecture A. Specific constraints associated with satellite networks
UE
NR-Uu
gNB-DU
F1
gNB-CU
NG(=N2/N3)
5GCN
N6
Data Network Compared to cellular systems, there are some design con-
straints that need to be addressed when considering satellite
3GPP NG-RAN (or NR radio access) architecture
network deployment scenarios:
Fig. 2: Comparison between the NG-RAN logical architecture • Propagation channel: the channel has a different multi-
and the satellite network physical architecture. path delay and Doppler spectrum model. However, for
narrowband signals and frequency bands below 6 GHz,
8
the time disparity may be ignored. Certain outdoor con- • Radio resource management adapted to network
ditions and line-of-sight operations are necessary for the topology: Unlike cellular systems where access control
UE communication via satellite. is typically located close to the UE, in satellite systems
• Frequency plan and channel bandwidth: The allocated access control is mostly located at the satellite base
spectrum to a satellite system is respectively 2 x 15 MHz station, gateway, or hub level, which may prevent an
(Uplink (UL) & Downlink (DL)) at S band and about optimal response time for access control. Hence, pre-
2 x 2.5 GHz for UL and 2 x 2.4 GHz for DL at Ka grants, Semi-Persistent Scheduling (SPS), and/or a grant-
band. In addition, satellite systems at S and Ka bands free access scheme would be beneficial.
use mostly circular polarizations. With frequency re-use • Terminal mobility: Very high speed UE with speeds of
and efficient spectrum allocation among different cells, up to 1,000 km/h need to be supported.
the maximum channel bandwidth per cell is assumed to
be respectively 2 x 15 MHz (UL and DL) at S band and
B. NR impacted areas and potential solutions
up to 2 x 2.4 GHz (UL and DL) at Ka band. However,
inter cell interference should be minimized. The aforementioned satellite related design constraints have
• Power limited link budget: Two main design aspects impacts on some features of NR as documented in 3GPP
need to be considered: TR 38.811 [10] and 3GPP TR 38.821 [14]. The following
points provide a brief description of how NR is impacted, and
– Maximizing the throughput for a given transmit Table IV summarizes the potential solutions to support NR in
power from the UE on the UL and from the satellite satellite networks .
on the DL.
• Handover paging: The fast movement of LEO satellites
– Maximizing the availability of the service under deep
and their many beams mean that UEs are only kept within
fading situations (typically between 20 and 30 dB in
a beam for a few minutes. The rapid change creates
Ka band for 99.95% availability).
problems for paging as well as for handovers for both
• Cell pattern generation: Satellites have larger cells stationary UEs and moving UEs. A handover has to be
compared to cellular networks, and the cells are moving executed quickly otherwise the UE may not make use of
in case of NGSO satellite. This creates a significant the satellite resources efficiently and may suffer a loss
difference in propagation delay between UE at the cell of data. With fixed tracking areas on the ground, there
edge and UE at the cell centre, and the difference in is no one-to-one correspondence between moving beams
propagation delay increases as the altitude of the satellite and fixed tracking areas or registration areas, which is
decreases. Accordingly, when the position of UE is not necessary for the paging process.
known by the network, contention-based channel access • Tracking Area (TA) adjustment: Moving satellites gen-
might be impacted. erate strong delay variations, and a fast change in the
• Propagation delay characteristics: Satellite systems overall distance of the radio link between the UE and BS
have much greater propagation delays than terrestrial via satellite. This delay largely exceeds the Transmission
systems, which can impact all round-trip signaling times, Time Interval (TTI) of NR, which is equal to or less
especially at access and transport (data transfer) levels. than 1 ms. Hence, the TA alignment is an important
• Mobility of the infrastructure’s transmission equip- feature of NR that will be impacted by the introduction
ment: For GSO satellites, the transmission equipment of satellites in 5G to ensure that all uplink transmissions
is quasi static with respect to the UE with only small are synchronized at gNB reception point.
Doppler effects. For NGSO satellites, by contrast, move • Synchronization in downlink: In order to access the
relative to the earth and create higher Doppler effects than 5G network, the UE has to detect the primary and
GSO systems. The Doppler depends on the frequency secondary synchronization signals. These synchronization
band and the relative satellite velocity with respect to signals allow time and frequency correction as well as
the UE. The Doppler effect will continuously modify the Cell ID detection. UE in cellular network has to get
carrier frequency, phase, and spacing. However, most of good one-shot detection probability at -6 dB received
the Doppler shift and variation rate can be compensated baseband Signal-to-Noise Ratio (SNR) condition with
by utilizing the predictable motion of satellites as well as less than 1% false alarm rate, with robustness against
UE location if known. initial frequency offset up to 5 ppm. It is expected that
• Service continuity between land-based 5G access and these requirements defined for terrestrial UE will be kept
non-terrestrial-based access networks: Different han- the same for SatNets UE. Even though the SNR level
dover triggering mechanisms are required in order to of satellite systems is typically in the range of -3 to 13
give preference to cellular communication. The handover dB SNR, the satellite movement creates a higher Doppler
procedure should support both regenerative and bent- shift, depending on the frequency band and the velocity
pipe satellites. In addition, it should handle handover of the satellite relative to the UE. However, this can be
preparation and failure while also supporting lossless compensated for at the demodulator.
handover. Handover can be due to intra-non-terrestrial • Hybrid Automatic Repeat Request (HARQ): The
network mobility as well as between non-terrestrial and HARQ process is a time-critical mechanism. In Sat-
cellular networks. Nets, the Round Trip Time (RTT) normally exceeds the
9
maximum conventional HARQ timers or the maximum and procedure design need to be modified.
possible number of parallel HARQ processes. This means • Access scheme (Time Division Duplex (TDD)/ Fre-
that simply extending the number of HARQ processes quency Division Duplex (FDD)): Most existing satellite
linearly to RTT might not be feasible for some UE due systems operate in the frequency bands designated for
to memory restrictions and the maximum possible parallel the FDD mode with a defined transmit direction. For
processing channels. Also, the impact of this delay has some frequency bands, the TDD mode is possible. When
to be considered by the gNBs on the number of their considering the TDD mode, a guard time is necessary
active HARQ processes. Although NR has extended the to prevent the UE from simultaneously transmitting and
number of HARQ processes in Rel. 15 to 16 processes, receiving. This guard time directly depends on the prop-
for SatNet NR the number of HARQ processes may need agation delay between the UE and gNB. This guard time
to be further extended flexibly according to the induced will directly impact the useful throughput and hence the
RTT delay. spectral efficiency.
• MAC/Radio Link Control (RLC) procedure: For LEO • Phase tracking reference signal (PT-RS): Phase vari-
satellite systems, the one way propagation delay changes ations in time domain can be caused by different phe-
continuously (e.g., 2-7 ms for 600 km orbit). The Auto- nomena, including the presence of phase noise, frequency
matic Repeat Request (ARQ) requires that the transmitted drifts due to Doppler shift, or due to insufficient fre-
packets be buffered and released only after the successful quency synchronization (e.g., residual CFO), etc. In NR,
receipt of an acknowledgement or until a time out. A PT-RS has been introduced to compensate for phase
larger transmission buffer is required due to the long RTT, errors. The PT-RS configuration in NR is very flexible
which also limits the number of retransmissions allowed and allows user-specific configurations depending on
for each transmitted packet. The ARQ transmit buffer scheduled MCS/bandwidth, UE RF characteristics, de-
size and retransmission mechanism must be designed for modulation reference signal configuration, waveform, etc.
the longest possible delay (i.e., at the lowest elevation). PT-RS configuration flexibility is beneficial for SatNets.
Scheduling mechanisms must be able to cope with the • Peak-to-average power ratio (PAPR): A key component
long RTT. in satellite payload architecture is a power amplifier. It ex-
• Physical layer procedure (Automatic Coding and hibits nonlinear behavior when operating near saturation
Modulation (ACM), power control): Due to the large in an effort to increase power efficiency. Nonlinear dis-
free-space loss and limited power available at the UE tortion causes constellation warping and clustering, thus
and satellites, the power margin is limited. Thus, only a complicating signal reception. PAPR is a measurement
limited amount of power control is available for satellite that determines the vulnerability of the transmitted signal
links. Due to the long delay in the loop, the power control to nonlinear distortion, where higher values indicate a
is not expected to track fast fading, but may be used to worse impact. In the NR downlink, Cyclic Prefix – Or-
track slower power variations. The slow reaction time, thogonal Frequency Division Multiplexing (CP-OFDM)
due to long RTT, is expected to impact the performance is used resulting in higher PAPR values compared with
of some physical layer procedures, particularly those with the underlying modulation in a single carrier. The dis-
close control loops, such as power control and ACM. tortion can be reduced by increasing the backoff of the
However, most control loops require some adjustments in amplifier operating point. But this reduces the amplifier
implementation, but not a fundamentally different design. efficiency accordingly. For satellite communications, low
• Time advance in Random Access Response (RAR) PAPR waveforms are desired.
message: Time advance mechanisms ensure that trans- • Protocols: Mapping is needed between the NG-RAN
missions from all UE operating in the same cell are logical architecture and the SatNet architecture (the two
synchronized when received by the same gNB. A time architectures are depicted in Fig. 2). Several mobility
advance command is provided to the UE in an RAR scenarios should be considered, specifically the mobility
message during initial access and later to adjust the induced by the motion of satellites, the motion of UEs
uplink transmission timing. The maximum value of the from one beam to another beam generated by the same
time advance command constrains the maximum distance satellite, the motion of UEs between beams generated by
between UEs and base station, which defines the allowed different satellites, and the motion of UEs between satel-
cell size. lites and cellular access. Location updating, paging, and
• Physical Random Access Channel (PRACH): It is handover RAN related protocols need to accommodate
necessary to consider the long RTT impact on PRACH. the extended delay of intra-satellite access mobility, the
For a given beam covering a cell, there is one common differential delay when mobility is between a satellite and
propagation delay for all served UE, and one relative a cellular network, and the mobility of the cell pattern
propagation delay for each served UE. If the common generated by NG-satellites.
propagation delay can be compensated, then the satellite
PRACH signal design will depend on the relative propa- VI. M ANAGEMENT AND O RCHESTRATION WITH
gation delay, which is limited to a TA range of up to 200 I NTEGRATED S ATELLITE C OMPONENTS IN A 5G N ETWORK
km in current NR specifications. However, when the TA Orchestration allows network services and resources to be
is of thousands of kilometers a satellite PRACH signal managed and controlled on an integrated basis and optimized.
10
3GPP 3GPP
Management Management
Satellite
Vertical / CSC
Satellite
Vertical / CSC
Non-3GPP
NR-RAT RAN CN
CN
UE Control Plane UE Control Plane
Terrestrial Terrestrial
RAT 3GPP RAN CN
CN
User Plane User Plane
3GPP RAN
Control/ User
Plane
Management
(a) Plane
(b)
Fig. 4: Reference management architectures for integrated satellite components as described in [13]. (a) Reference architecture
for the management of a 3GPP RAN integrating a satellite NR-RAT with a terrestrial RAT. (b) Reference architecture for the
management of a non-3GPP satellite RAN integrated in a 5G network.
Thus, time spent on processes is shortened by rapidly and VII. OTHER S TANDARDIZATION O RGANIZATIONS
flexibly allocating network components and resources. In
a) International Telecommunication Union (ITU) Stan-
2019, the 3GPP TSG SA started a study on management and
dardization: In ITU terminology, the 5G system is called
orchestration aspects with integrated satellite components in a
IMT-2020. The IMT-2020 network architecture is envisioned
5G network. The main objective was to study business roles
to be access network-independent and with a core network
as well as service, network management, and orchestration of
common to Radio Access Technology (RAT) for IMT-2020,
a 5G network with integrated satellite components. The scope
as well as existing fixed and wireless networks. The IMT-2020
of the study covered both NTN RAN-based satellite access,
core network control mechanisms will be decoupled from the
non-3GPP defined satellite access, as well as backhaul aspects.
access network technologies. The IMT-2020 network should
The study outcome was presented in TR 28.808 [13] which
support new RATs for IMT-2020, evolved IMT-advanced
included potential requirements and solutions to integrate
RATs, satellite networks, fixed broadband network access, and
satellites in 5G networks, such as network slice management
Wireless Local Area Network (WLAN) access networks. In
and management and monitoring of gNB components.
July 2020, 3GPP 5G has been formally endorsed as ITU IMT-
2020 5G standard. The M family of ITU-R recommendations
The study presented two reference management architec-
refers to 5G systems. A series of ITU-R recommendations
tures for integrated satellite components, as shown in Figure
(standards) for the satellite components for IMT have been
4. The first reference architecture was for the management
already developed, which includes the integration of terrestrial
of a 3GPP RAN integrating a satellite NR-RAT with a
and satellite mobile communication systems [20]:
terrestrial RAT. The second reference architecture was for the
management of a non-3GPP satellite RAN integrated in a 5G • Rec. ITU-R M.1167 - “Framework for the Satellite Com-
network. The potential requirements for the management and ponent of International Mobile Telecommunications-2000
orchestration of integrated satellite components in a 5G net- (IMT-2000)”.
work are shown in Figure 5. The requirements are presented in • Rec. ITU-R M.818 - “Satellite Operation within Interna-
three categories: (1) network slice management requirements; tional Mobile Telecommunications-2000 (IMT2000)”.
(2) management of satellite components; and (3) monitoring • Rec. ITU-R M.1850 - “Detailed Specifications of the Ra-
lites mainly comes from LEO/MEO scenarios where gNB Satellite Mobile Communication Systems”.
components, such as gNB-DU, are located onboard satellites • Rec. ITU-R M.2014 - “Global Circulation of IMT-2000
and therefore move faster than the earth. Other enhancements Satellite Terminals”.
are required due to the long delays and RTT that impact some On the other hand, ITU-R Working Party 4B (WP 4B)
of the monitoring functionality and key performance indicators has carried out studies on performance, availability, air in-
of 5G networks. The study concluded that the concepts of terfaces and earth-station equipment of satellite systems in
Self-Organizing Networks (SONs) for 5G would need to be Fixed Satellite Services (FSSs) as well as land, maritime, and
enhanced to support mobile non-terrestrial gNBs. Although aeronautical Mobile Satellite Services (MSSs), and Broadcast
efficient network management is essential in future integrated Satellite Services (BSSs) [21]. This group has given special
networks in order to fully utilize the available network re- consideration to the study of Internet Protocol (IP) related
sources, the standardization work on NTN management is system aspects and performance, and they have developed
nevertheless quite limited within the 3GPP working groups. new and revised recommendations and reports on IP over
11
Potential requirements for the management and orchestration of integrated satellite components in a 5G network
• In a 5G network with satellite access, the satellite components shall have the capability to provide NSIs and NSSIs for each of the NOP sharing the satellite access network.
• In a 5G network with satellite access, each of the NOPs sharing the satellite access shall have the capability to create, manage, monitor, activate, deactivate and terminate NSIs and NSSIs.
Network slice instance(s) associated with both a satellite RAN and a terrestrial RAN
• A single NSI with both satellite access and terrestrial access, shall have the possibility to set separate network slice service requirements for the Satellite RAN and Terrestrial RAN domains.
Management
components
of satellite
• In a 5G network integrating a NGSO component with split gNBs, it shall possible to monitor the packet delay KPIs.
components
Monitoring of average delay on DL-Air Interface with MEO or GEO satellite components
• In a 5G network integrating a MEO or GEO satellite components, it shall be possible to monitor the average delay on the DL-Air Interface.
Multi-RAT load-balancing associated with both a satellite RAN and a terrestrial RAN
• Multi-RAT load-balancing shall be provided to assure optimal resource usage of the satellite RAN and terrestrial RAN domains.
• Multi-RAT load-balancing shall be provided to guarantee service continuity (reliability) in mobile UEs.
Fig. 5: Potential requirements for the management and orchestration of integrated satellite components in a 5G network.
satellite to meet the increasing demands of satellite links to •Regulatory frameworks for sharing between GSO and
carry IP traffic. This group cooperates closely with the ITU non-GSO satellite systems in the 50/40 GHz range. Also
Telecommunication Standardization Sector. sharing between GSO FSS, BSS & MSS and non-GSO
b) World Radio Conference (WRC): World radio com- FSS satellite systems.
munication conferences are held every three to four years. • Deployment process should be based on milestone to
WRC reviews and, if necessary, revises the Radio Regula- avoid spectrum warehousing by large non-GSO satellite
tions, the international treaty governing the use of the radio- constellations.
frequency spectrum and the geostationary-satellite and non- • One of the new regulations adopted is that non-GSO
geostationary-satellite orbits. Revisions are made on the basis systems have to deploy 10% of the constellation within
of an agenda determined by the ITU Council, which takes two years, 50% within five years and the full deployment
into account recommendations made by previous world radio within seven years.
communication conferences. WRC-19 took place in Sharm El. c) Internet Engineering Task Force (IETF): IETF is
Sheikh, Egypt on October 28 to November 22, 2019, and 3,540 a large open international community of network designers,
delegates from 165 countries attended. The members took the operators, vendors, and researchers concerned with the evolu-
following decisions [22]: tion of Internet architecture and the smooth operation of the
• Embedded SIM (ESIM) – expected to provide reliable Internet [23]. Under IETF, the Transport and Services Area
and high bandwidth internet services to aircraft, ships, covers a range of technical topics related to data transport in
and land vehicles. the Internet, such as protocol design and maintenance at Layer
• Resolution lays out technical and regulatory conditions 4 (e.g., Transmission Control Protocol (TCP), User Datagram
for three types of ESIM communicating with a GSO FSS Protocol (UDP), and Stream Control Transmission Protocol
space stations within the frequency band 17.7-19.7 GHz (SCTP)), congestion control and (active) queue management,
(space to earth) and 27.5-29.5 GHz (earth to space). and QoS and related signaling protocols. The transport area
12
subcommittee of the IETF are working with network operators and Telecommunications Administrations. The ECC Working
to standardize a multi-path deployment scheme that includes Group Frequency Management (WG FM) is responsible for
fixed gateways and satellites for the backhauling of the 5G developing strategies, plans, and implementation advice for
network. With Multi-Path TCP (MPTCP), communication ser- the management of the radio spectrum. WG FM44 deals with
vice providers can extend the coverage and the bandwidth of satellite communications in particular [26].
5G services. This will address the issues where in some areas g) National Institute of Information and Communica-
the fixed network is not able to deliver enough bandwidth for tion Technology (NICT) Japan: The Study Group on the
the backhauling. The MPTCP working group has developed Integration of Satellite Communications and 5G/Beyond 5G
mechanisms that add the capability of simultaneously using of the National Institute of Information and Communications
multiple paths to a regular TCP session without making any Technology (NICT) Japan identified the use cases that Japan
assumption about the support of the communicating peers. In would need by 2040 [27]. The themes of these use cases were
this context, satellite networks can be considered as one of the the Internet of Things, smart cities, maritime and aviation,
options when more than one path is used for backhauling. transportation infrastructures, and responding to emergency
d) European Telecommunications Standards Institute disasters. The use cases require the integration of satellite
(ETSI): ETSI has investigated a number of component tech- networks and 5G+ networks. The report produced by NICT
nologies that will be integrated into future 5G systems, highlighted some standardization requirements (e.g., standard
such as Network Function Virtualization (NFV), Mobile Edge communication protocols for every layer) that need to be
Computing (MEC), Millimeter Wave Transmission (mWT), considered in order to realize the identified use cases.
and Non-IP Networking (NIN) [24]. The objective of ETSI h) IEEE International Network Generation Roadmap
activities is to define the end-to-end SatCom system that can be (INGR) - Satellite Work Group: IEEE INGR-Satellite WG
fully integrated in 5G. The standardization work is undertaken produced two editions of its roadmap report in 2020 and 2021
by the following technical committees: [28] [29]. The 2021 Edition of the INGR Satellite Working
• Satellite Earth Stations and Systems (TC-SES): Defines Group Report discussed several topics related to satellites in
all aspects related to satellite earth stations and sys- future networks, including applications and services, reference
tems. Within TC-SES, the SCN working group (Satel- architectures, new MIMO-based PHY, antenna and payload,
lite Communication and Navigation) covers radio and machine learning and artificial intelligence, edge computing,
transmission aspects related to fixed and mobile satellite QoS/Quality of Experience (QoE), security, network man-
systems operating in any bands allocated to FSS, MSS, agement and standardization. Under each topic, the related
or global navigation satellite systems operating in any challenges, enablers, and potential solutions were highlighted.
bands allocated to RDSS. ETSI TC-SES has established
a working relationship with 3GPP. This has allowed the
development of standards for mobile satellite as well as A. Satellite and aviation standardization
broadband satellite multimedia systems, which are mainly a) Digital Video Broadcasting (DVB): This industry
based on 3GPP system architecture and radio protocols. forum was established initially to create a set of standards
• Network Function Virtualization (ISG-NFV): Produces for the TV broadcasting industry via terrestrial, satellite, or
the technical specifications for the virtualization of net- cable networks. DVB has successfully defined the DVB-
work functions. S2 broadcast channel, which is widely adopted around the
• Open Source MANO (OSM): Develops a software refer- world. A Return Channel (DVB-RCS/2) has been defined to
ence implementation of ETSI Management and Orches- support broadband telecommunication services via satellite.
tration (MANO). The DVB technical specifications were published by ETSI
• Multi-Access Edge Computing (TC-MEC): Produces the after being reviewed in the TC-BROADCAST. However, DVB
technical specifications for realizing Multi-Access Edge technical specifications have often been associated with some
Computing (MEC) in the context of content delivery proprietary features at the level of architecture, protocol stack,
(multicasting and caching). and radio access. This has resulted in interoperability prob-
e) 5G Infrastructure Public Private Partnership (5G lems among satellite access networks from different solution
PPP): The 5G PPP is a joint initiative between the European vendors, and it has led to a fragmented SatCom market.
Commission and the European ICT industry. The 5G PPP will b) Consultative Committee for Space Data Systems
deliver solutions, architectures, technologies, and standards (CCSDS): CCSDS was formed in 1982 by the major space
for the ubiquitous 5G communication infrastructures. In June agencies of the world to provide a forum for discussing com-
2021, The 5G PPP infrastructure association published a new mon problems in the development and operation of space data
white paper, “Vision and Societal Challenges Work Group,” systems [30]. CCSDS has been actively developing standards
which described the European vision for the 6G network for data-systems and information-systems to promote interop-
ecosystem [25]. This white paper covered key areas related erability and cross support among cooperating space agencies,
to 6G research from a technical, societal, policy, and business to enable multi-agency spaceflight collaboration (both planned
perspective, providing a vision for future mobile networks. and contingency) and new capabilities for future missions. The
The paper dedicated one section to non-terrestrial networks. CCSDS standardization work reduces the cost of spaceflight
f) Electronic Communication Committee (ECC): ECC missions by allowing cost sharing between agencies and cost-
is an organization of the European Conference of Postal effective commercialization. CCSDS has six technical areas
13
with twenty-three working groups. The working body respon- time. Managing the handover of thousands of users simulta-
sible for defining communications standards is the Space Link neously or semi-simultaneously using conventional handover
Service, composed of six working groups. It defines two management schemes will create huge network loads. New
main links between earth and space probes: telemetry and handover management schemes are required to deal with this
telecommand. issue in 6G LEO SatNets.
c) Airlines Electronic Engineering Committee (AEEC): For mobility management in IP-based networks, IETF in-
Develops engineering standards and technical solutions for troduced a number of protocols, such as Mobile Internet
avionics, networks, and cabin systems that foster increased Protocol version 6 (MIPv6) and Proxy Mobile Internet Pro-
efficiency and reduced life cycle costs for the aviation com- tocol version 6 (PMIPv6). However, such protocols were not
munity [31]. Their standardization work is overseen by the designed to deal with the high topology change rate in SatNets,
subcommittee Network Infrastructure and Security, which where everything is moving including the gNB (LEO satellite
aims to develop standards for IP connectivity and security base station). A number of approaches have been proposed
for aircraft and to enable fleet-wide solutions based on open to address this problem [32]. Nevertheless, the concept of
standards for lower development costs, increased flexibility, separating control plane and data plane of Software Defined
higher reliability, reduced complexity, longer lifespans, and Network (SDN) is a promising approach to efficiently manage
ease of configurability and maintenance. ARINC is AEEC SatNet topology.
Project Paper 848 - “Secure Broadband IP Air-Ground In- The fast-moving footprint of LEO satellites affects the
terface (SBAGI),” which was intended to define a method for paging procedure, which is primarily related to the tracking
secure communications interface between IP networks con- area management. The tracking area is the satellite coverage
tained within an aircraft system and within a ground network area (footprint); it can be fixed or moving. Although the
hosted by the aircraft original equipment manufacturer, airline moving tracking area accommodates the LEO satellite moving
or a 3rd party. The ARINC Project Paper 848 standardized footprint, it results in high paging loads that are difficult
this interface at the network level while taking into account to manage by the network. In addition, supporting dual-
the overall security context. connectivity and vertical handovers in future LEO SatNets
requires novel mechanisms to provide seamless mobility in
VIII. S TANDARDIZATION FOR 6G S ATELLITE integrated 6G networks and to improve global network cover-
C OMMUNICATIONS N ETWORKS age and service.
Most of the standardization work carried out by 3GPP and
other standardization organizations focuses on the physical and
MAC layers. Consideration has also been given to defining B. Routing
satellite use cases and architectural options in the context of
integrated satellite 5G networks. The following subsections One very important characteristic of LEO mega-
highlight several issues that need to be considered in stan- constellations is the ability of satellites to form networks and
dardization work in order to achieve the complete integration communicate with each other through ISLs [33]. Due to the
of satellite and terrestrial 6G networks. frequent topology changes in an LEO SatNet, ISLs have a
limited lifetime. In addition, some ISLs may get congested
A. Mobility management due to high traffic loads at certain partitions of the SatNet.
Moreover, as LEO SatNets are expected to serve different
LEO satellites provide shorter propagation delays and higher
types of applications, there are certain QoS requirements (e.g.,
data rates than GEO satellites. However, these advantages
packet delivery delay, packet delivery ratio) that need to be
come with the price of frequent handover and topology
met for each type of applications. Therefore, successful data
changes, which yields a time-varying communication channel.
delivery will require robust routing schemes that can fulfil the
Handovers in LEO SatNets are of three types:
QoS requirements of each application type and adapt to the
• Intra-satellite handovers, which occur between satellite unique characteristics of LEO SatNets. For example, delay-
beams. tolerant routing is suitable for delay-tolerant applications,
• Inter-satellite handovers, which occur between satellites. while multi-path routing is required to support applications
• Inter-access network handovers (also known as vertical with high bandwidth requirements. Thus, it is crucial to
handovers), which occur either between satellites belong- develop standard routing protocols that adapt to the SatNet
ing to different access networks or from a SatNet to an dynamic environment and satisfy the various user application
ariel network or terrestrial network (or vice versa) in requirements. Standards should support interoperability
integrated terrestrial-NTN systems. among the different satellite constellations and operators.
In 6G future networks, LEO SatNets will not only serve Moreover, cross network routing (i.e., across satellite, aerial,
rural or remote areas but will also provide communication and terrestrial networks) should be considered to achieve the
services and coverage in urban and highly populated areas. full integration in 6G. To support efficient routing, topics such
Such a scenario will lead to thousands of UE being connected as resource allocation, network monitoring, and congestion
to an LEO satellite and this large group of users will need to control should be considered as part of the standardization
go through a frequent handover process at almost the same work.
14
[3] “Study on Scenarios and Requirements for Next Generation Access [28] IEEE International Network Generations Roadmap - Satellites, “ An
Technologies,” document TR 38.913, V14.2.0, Release 14, Third Gen- IEEE 5G and Beyond Technology Roadmap-1st edition,” 2020.
eration Partnership Project (3GPP), 2017. [29] ——, “ An IEEE 5G and Beyond Technology Roadmap-2st edition,”
[4] F. Rinaldi, H.-L. Maattanen, J. Torsner, S. Pizzi, S. Andreev, A. Iera, 2021.
Y. Koucheryavy, and G. Araniti, “Non-Terrestrial Networks in 5G & [30] “The Consultative Committee for Space Data Systems (CCSDS),” https:
Beyond: A Survey,” IEEE Access, vol. 8, pp. 165 178–165 200, 2020. //public.ccsds.org/about/default.aspx, accessed: 2021-07-31.
[5] S. C. Burleigh, T. De Cola, S. Morosi, S. Jayousi, E. Cianca, and [31] “Airlines Electronic Engineering Committee (AEEC),” https://www.
C. Fuchs, “From Connectivity to Advanced Internet Services: A Com- aviation-ia.com/articles/aeec-sets-standards, accessed: 2021-07-31.
prehensive Review of Small Satellites Communications and Networks,” [32] T. Darwish, G. Karabulut Kurt, H. Yanikomeroglu, G. Lamontagne, and
Wireless Communications and Mobile Computing, vol. 2019, 2019. M. Bellemare, “Location Management in IP-based Future LEO Satellite
[6] A. Guidotti, S. Cioni, G. Colavolpe, M. Conti, T. Foggi, A. Mengali, Networks: A Review,” arXiv preprint arXiv:2101.08336, 2021.
G. Montorsi, A. Piemontese, and A. Vanelli-Coralli, “Architectures, [33] Q. Chen, G. Giambene, L. Yang, C. Fan, and X. Chen, “Analysis of
Standardisation, and Procedures for 5G Satellite Communications: A Inter-Satellite Link Paths for LEO Mega-Constellation Networks,” IEEE
Survey,” Computer Networks, vol. 183, p. 107588, 2020. Transactions on Vehicular Technology, vol. 70, no. 3, pp. 2743–2755,
[7] A. Guidotti, A. Vanelli-Coralli, M. Conti, S. Andrenacci, S. Chatzinotas, 2021.
N. Maturo, B. Evans, A. Awoseyila, A. Ugolini, T. Foggi et al., “Ar- [34] “ETSI Experiential Networked Intelligence Industry Specifi-
chitectures and Key Technical Challenges for 5G Systems Incorporating cation Group (ENI ISG),” https://www.etsi.org/technologies/
Satellites,” IEEE Transactions on Vehicular Technology, vol. 68, no. 3, experiential-networked-intelligence, accessed: 2021-07-31.
pp. 2624–2639, 2019. [35] Y. Wang, R. Forbes, U. Elzur, J. Strassner, A. Gamelas, H. Wang, S. Liu,
[8] X. Lin, S. Rommer, S. Euler, E. A. Yavuz, and R. S. Karlsson, “5G L. Pesando, X. Yuan, and S. Cai, “From Design to Practice: ETSI
from Space: An Overview of 3GPP Non-Terrestrial Networks,” arXiv ENI Reference Architecture and Instantiation for Network Management
preprint arXiv:2103.09156, 2021. and Orchestration Using Artificial Intelligence,” IEEE Communications
[9] O. Kodheli, E. Lagunas, N. Maturo, S. K. Sharma, B. Shankar, J. F. M. Standards Magazine, vol. 4, no. 3, pp. 38–45, 2020.
Montoya, J. C. M. Duncan, D. Spano, S. Chatzinotas, S. Kisseleff, [36] “Technical Specification Group Services and System Aspects, Telecom-
J. Querol, L. Lei, T. X. Vu, and G. Goussetis, “Satellite Communi- munication Management, Self-Organizing Networks (SON) Concepts
cations in the New Space Era: A Survey and Future Challenges,” IEEE and requirements,” document TS 32.500, V16.0.0, Release 16, Third
Communications Surveys and Tutorials, vol. 23, no. 1, pp. 70–109, 2021. Generation Partnership Project (3GPP), 2020.
[10] “Technical Specification Group Radio Access Network, Study on New [37] T. Darwish, G. Karabulut Kurt, H. Yanikomeroglu, G. Senarath, and
Radio (NR) to Support Non-Terrestrial Networks (NTN),” document P. Zhu, “A Vision of Self-Evolving Network Management for Future
TR 38.811, V15.4.0, Release 15, Third Generation Partnership Project Intelligent Vertical HetNet,” IEEE Wireless Communications, vol. 28,
(3GPP), 2020. no. 4, pp. 96–105, 2021.
[11] “Technical Specification Group Services and System Aspects, Study on [38] A. Farajzadeh, M. G. Khoshkholgh, H. Yanikomeroglu, and O. Ercetin,
Using Satellite Access in 5G, Stage 1,” document TR 22.822, V16.0.0, “Self-Evolving Integrated Vertical Heterogeneous Networks,” arXiv
Release 16, Third Generation Partnership Project (3GPP), 2018. preprint arXiv:2106.13950, 2021.
[12] “Technical Specification Group Services and System Aspects, Study [39] Y.-C. Liang, J. Tan, H. Jia, J. Zhang, and L. Zhao, “Realizing Intelligent
on Architecture Aspects for Using Satellite Access in 5G,” document Spectrum Management for Integrated Satellite and Terrestrial Networks,”
TR 23.737, V17.2.0, Release 17, Third Generation Partnership Project Journal of Communications and Information Networks, vol. 6, no. 1, pp.
(3GPP), 2021. 32–43, 2021.
[13] “Technical Specification Group Services and System Aspects, Study [40] K. Tekbıyık, G. Karabulut Kurt, A. Rıza Ekti, A. Görçin, and
on Management and Orchestration Aspects with Integrated Satellite H. Yanikomeroglu, “Reconfigurable Intelligent Surfaces Empowered
Components in a 5G Network,” document TR 28.808, V17.0.0, Release THz Communication in LEO Satellite Networks,” arXiv e-prints, p.
17, Third Generation Partnership Project (3GPP), 2021. arXiv:2007.04281, 2020.
[41] A. U. Chaudhry and H. Yanikomeroglu, “Free Space Optics for Next-
[14] “Solutions for NR to Support NTN,” document TR 38.821, V16.1.0,
Generation Satellite Networks,” IEEE Consumer Electronics Magazine
Release 16, Third Generation Partnership Project (3GPP), 2021.
(Early Access), 2020.
[15] “3GPP SA WG, Update to 5GSAT: Integration of Satellite Access in
5G,” Release 16,Third Generation Partnership Project (3GPP), 2020.
[16] “Advanced Plans for 5G,” https://www.3gpp.org/news-events/
2210-advanced5g, accessed: 2021-07-31.
[17] “Telesat,” https://www.telesat.com/, accessed: 2021-09-30.
[18] O. Kodheli, A. Guidotti, and A. Vanelli-Coralli, “Integration of Satellites
in 5G through LEO Constellations,” in IEEE Global Communications
Conference (GLOBECOM), 2017, pp. 1–6.
Tasneem Darwish is currently a postdoctoral fellow
[19] “Technical Specification Group Radio Access Network, NR, Study
at the Department of Systems and Computer Engi-
on Integrated Access and Backhaul,” document TR 38.874, V16.0.0, neering, Carleton University, Canada. She received
Release 16, Third Generation Partnership Project (3GPP), 2019. the MSc. degree with merit in Electronics and Elec-
[20] “ITU M Series,” https://www.itu.int/rec/R-REC-M/en, accessed: 2021- trical Engineering from the University of Glasgow,
07-31. UK, in 2007 and her Ph.D. degree in Computer
[21] “ITU Working Party 4B (WP 4B),” https://www.itu.int/en/ITU-R/ Science from Universiti Teknologi Malaysia (UTM),
study-groups/rsg4/rwp4b/Pages/default.aspx, accessed: 2021-07-31. Malaysia, in 2017. From 2017 to 2019, Tasneem was
[22] “World Radio communication Conferences (WRC),” https://www.itu.int/ a postdoctoral fellow at UTM . From 2019 to 2020,
en/ITU-R/conferences/wrc/Pages/default.aspx, accessed: 2021-07-31. she was a research associate at Carleton University,
[23] A. Ford, C. Raiciu, M. J. Handley, O. Bonaventure, and C. Paasch, Canada. In 2020, Tasneem started working as a
“TCP Extensions for Multipath Operation with Multiple Addresses,” postdoctoral fellow at Carleton University on a collaborative project with
RFC 8684, Mar. 2020. [Online]. Available: https://rfc-editor.org/rfc/ MDA Space to investigate mobility management in future LEO satellite
rfc8684.txt networks. She is the recipient of the UTM Alumni Award for Science and
[24] “European Telecommunications Standards Institute (ETSI),” https:// Engineering in 2017. She was awarded the Malaysia International Scholarship
www.etsi.org/standardsPre-defined%20Collections, accessed: 2021-07- (MIS) from 2013 to 2016. Her current research interests include mobility
31. management in future LEO satellite networks, edge/fog computing and data
[25] “5G Infrastructure Public Private Partnership (5G PPP),” https://5g-ppp. offloading in HAPS, vehicular ad hoc networks, and intelligent transportation
eu/european-vision-for-the-6g-network-ecosystem/, accessed: 2021-07- systems. Tasneem is a senior IEEE member and an active reviewer for
31. several IEEE journals such as IEEE Internet of Things, IEEE Access, IEEE
[26] “Electronic Communications Committee’s (ECC),” https://cept.org/ecc/ Transactions on Vehicular Technology, and IEEE Transactions on Intelligent
groups/ecc/wg-fm/fm-44/client/introduction/, accessed: 2021-07-31. Transportation Systems.
[27] National Institute of Information and Communications Technology
(NICT), “ The Integration of Satellite Communications and 5G/Beyond
5G,” 2021.
16
Gunes Karabulut Kurt is currently an Associate Guillaume Lamontagne received the B.Eng. and
Professor of Electrical Engineering at Polytechnique M.Eng. degrees from the École de Technologie
Montréal, Montreal, QC, Canada. She received the Supérieure (ÉTS), Montréal, QC, Canada, in 2007
B.S. degree with high honors in electronics and and 2009 respectively. His experience in satellite
electrical engineering from the Bogazici University, communications started through internships and re-
Istanbul, Turkey, in 2000 and the M.A.Sc. and the search activities with the Canadian Space Agency
Ph.D. degrees in electrical engineering from the (CSA), in 2005, and the Centre national d’études
University of Ottawa, ON, Canada, in 2002 and spatiales (Cnes), France, in 2006 and 2008. He
2006, respectively. From 2000 to 2005, she was joined MDA in 2009 and held various communica-
a Research Assistant at the University of Ottawa. tion systems engineering and management positions
Between 2005 and 2006, Gunes was with TenXc before being appointed as the Director of Technol-
Wireless, Canada. From 2006 to 2008, she was with Edgewater Computer ogy, Payloads, in 2019. Through this role, he is leading MDA’s Research and
Systems Inc., Canada. From 2008 to 2010, she was with Turkcell Research Development activities for satellite communications as well as establishing
and Development Applied Research and Technology, Istanbul. Gunes has been the related long term development strategy.
with Istanbul Technical University since 2010, where she is currently on a
leave of absence. She is a Marie Curie Fellow and has received the Turkish
Academy of Sciences Outstanding Young Scientist (TÜBA-GEBIP) Award in
2019. She is an Adjunct Research Professor at Carleton University. She is
also currently serving as an Associate Technical Editor (ATE) of the IEEE
Communications Magazine and a member of the IEEE WCNC Steering Board.
TABLE IV: Impacted NR areas and potential solutions to support satellite networks.
Satellite Effects Impacted NR Potential adaptations Potential areas of impact to be further studied
network features
specifics
Motion Moving Handover pag- The NR UE may need to be capable of reporting its geo-location to the Taking advantage of the knowledge of the UE
of the cell ing satellite RAN. The ephemeris information of the NGEO satellites can be used location and satellite ephemeris information to
space pattern to determine the satellite or beam footprint. Thus, the network can determine adapt the NR handover and paging protocols needs
vehicles which satellite/beam covers that location of a UE at any given time and for further study. The NR beam management might
(espe- which duration. assume the same frequency on the adjacent beams,
cially for but on the same satellite adjacent beams may
Non- use different frequencies or different polarization,
GEO- which would require the NR beam management
based procedures to be modified.
access Delay Tracking Knowing the satellite orbits and UE position, the delay variation is quite Solutions need to be studied to ensure the align-
network) variation area (TA) predictable. Also, the transmission timing of the UE has to be adjusted over ment of uplink signals over the satellite links to
adjustment the borders of individual TTIs. overcome the predictable delay of SatNets.
Synchronization To meet the condition of 5ppm, the satellite altitude has to be above 13,000 km. In case the aforementioned conditions are not met,
Doppler
in downlink For the Doppler shift amplitude to be compensated, it has to be less than 48 further studies are required to accommodate the
kHz for S band and 480 kHz for Ka band high Doppler shift during the cell synchronization
procedure in SatNets.
Hybrid Dealing with a higher number of parallel HARQ processes and its feasibility The impacts of SatNet delays on the existing
Long
Altitude automatic for non-terrestrial networks have been addressed in [18] [14]. HARQ-supporting mechanisms need to be studied.
latency
repeat request In particular, what needs to be examined is the
(HARQ) impact of long RTT delays on NR HARQ opera-
tions. Also the impact on UEs and serving satellite
gNBs should be considered when the number of
HARQ processes is either extended to satisfy high
reliability scenarios or limited/disabled for longer
SatNet delays.
MAC/Radio NR UE and base stations must size their transmission buffers and the retrans- No further study required.
Link Control mission time-out mechanisms according to the longest RTT to be anticipated.
(RLC) The number of retransmissions allowed before a packet is dropped from the
Procedures retransmission buffer may also be adjusted. UL scheduling delay parameters are
expected to be redefined to accommodate the RTT of the associated deployment
scenario.
Physical layer For LEO satellites, ACM may also be used to adapt to the large variation of free Further studies are needed to define the required
Procedures space loss. The variation is sufficiently slow compared to the 20 ms worst case margin for power and ACM control loops to cope
(ACM, power RTT. ACM should also be able to react to shadowing fades to a large extent, with the long RTT.
control) but still unable to follow fast fading.
Cell size Differential Time advance For satellites, it is expected that a time advance mechanism will need to be Further study is required as time advance should
delay in random modified to compensate for the propagation delay. compensate for differential delay/distance.
access
response
(RAR)
message
Physical The current window for the PRACH response in NR cannot cover the long The solution to handling long RTT with the con-
random access RTT. Therefore, the random access response window length in NR should be sideration of power saving at the UE side needs to
channel revisited to accommodate the RTT of satellites. However, extending the window be further studied. Further studies should be done
(PRACH) size in the existing procedure introduces unnecessary UE monitoring intervals. if a new random access preamble format is needed
For satellites, the current NR preamble format needs to be extended considering for satellites.
different footprint of satellite cells.
Duplex Regulatory Access scheme When considering satellite networks, this guard time should equal the round trip The applicability of the duplexing mode TDD or
mode con- (time division delay. Although this guard time may be acceptable in the case of LEO access FDD depends on the regulations (ITU-R and/or
straints duplex (TDD)/ system, there is a need to deal with the variable delay. national) associated with the targeted spectrum. In
frequency case the regulations allow it, TDD mode can be
division considered for LEO-satellite-based access network
duplex (FDD)) with some required NR modifications.
Satellite Phase Phase tracking PTRS is needed in NR supporting SatNets for phase error compensation. Typical Considering the high speed of LEO satellites,
Payload noise im- reference sig- phase noise masks of state-of-the-art bent pipe of satellite payloads in non- some further investigation is needed in the case of
performance pairment nal (PT-RS) terrestrial networks can be efficiently compensated by the current NR design important Doppler shifts and/or residual CFO, or
in the absence of important Doppler shifts and/or residual CFO at a carrier in the presence of the specific phase noise masks
frequency of up to 30 GHz. It has been proposed that eMBB should be supported of on-board payloads that are significantly differ-
by SatNets and the modulation order might not always be low. PT-RS is needed ent from the ones considered in cellular networks
in SatNets at high carrier frequencies, where the maximum received SNR is so far, or with very large channel bandwidths.
limited by the phase noise.
Back-off Peak-to- The use of CP-OFDM for downlink does not restrict NR operation in SatNets, PAPR reduction techniques of CP-OFDM signal
average power but it may affect the system performance. When a satellite transponder is for downlink would be beneficial for optimizing
ratio (PAPR) sufficiently wide and powerful to accommodate more than one FDM carrier, each the capacity of non-terrestrial networks and there-
of which is a different NR CP-OFDM signal, the satellite amplifier is backed off fore could be considered in future studies.
to minimize the intermodulation between these FDM carriers within the same
transponder. As a result, the distortion introduced is small. For communicating
with small UEs, the satellite amplifier is used to send only one NR CP-OFDM
downlink. It is highly desirable to operate the amplifier with as small an output
power backoff (OBO) as possible. But, due to the higher PAPR of CP-OFDM
signal, sufficient OBO is necessary. To close the link, it may be necessary to
reduce the size of CP-OFDM carrier or to operate the CP-OFDM carrier with
a lower modulation and coding mode. Either way, the forward link capacity is
reduced significantly. On the uplink, DFT-spread-OFDM might be beneficial.
Network RAN Protocols In [10] three mapping options are described that prevent the need to create new The N1/N2/N3 and GTP-based F1 interface pro-
architec- Mapping interfaces or reference points, as described in Figure 3. The timers associated to tocols may need to be adapted to accommodate
ture the protocols transported over the feeder and service links (e.g., F1, N1, N2 and the satellite feeder link characteristics (long de-
N3 reference points) may need to be extended. In the processed payload option, lay, BER). Depending on the SatNets deployment
it is recommended that the satellite implements some kind of “intermediate” scenarios, other impacts on NR specification may
node, which could be based on the outcomes of 3GPP TR 38.874 [19], and see have to be considered, such as location update,
how it can be applicable and what areas of impact it may create. Location update, paging, and handover RAN related protocols and
paging, and handover RAN related protocols need to be adapted depending on handling of network identifies.
tracking area design. Also, handling of network identities needs to be adapted.