Special Issue of IEEE

Advanced Solutions for 6G satellite

Systems Workshop

Reference Architectures for Enabling Integrated

Satellite-6G Applications and Services
Debabrata Dalai Sarath Babu B.S. Manoj
Indian Institute of Space Science Iowa State University, Ames, USA Indian Institute of Space Science
and Technology, Trivandrum, India 50011 and Technology, Trivandrum, India
695547 [email protected] 695547
[email protected] [email protected]

ABSTRACT
Reference Architectures (RAs) play an important role in the integration of 6G terrestrial
and satellite networks. In this paper, we present the essence of the reference
architectural roadmap as per the Edition-3 document of the Satellite Working Group of
IEEE Future Networks Initiative (FNI). We focus on an integrated virtualized 6G-satellite
architecture. Further, we present one of the case studies, Space Based Hosting Service
(SBHS) approach with simulation results. SBHS approach deploys the content-servers
in LEO satellite to achieve the low-latency service requirements. The architecture of
SBHS is a special case over Reference Architecture-3. We achieved minimum average
end-to-end latency of 7.75ms for the geographical area covering India by the proposed
SBHS approach.
Index Terms—6G-satellite integration, reference architecture, satellite network,
virtualization, MANO, satellite edge computing, content server.

1. INTRODUCTION
Recent studies in [1] [2] [3] estimate that approximately 37% of the world’s population is
unconnected to high-speed Internet access. Terrestrial networks cannot guarantee
access to the Internet to the users on aircraft or high-speed trains, highways, ships, and
remote rural areas due to lack of infrastructure deployment. However, the nodes in Non-
Terrestrial Networks (NTNs) such as Low Earth Orbits (LEOs), Medium Earth Orbits
(MEOs), Geostationary Equatorial Orbits (GEOs) satellites, Unmanned Aerial Vehicles
(UAVs), and High-Altitude Platforms (HAPs) can help in providing Internet to such
scenarios.
The integration of 6th Generation (6G) wireless communication systems with NTNs has
brought a paradigm shift in employing technologies such as softwarization and
virtualization. The elements of NTNs primarily include satellites in LEO, MEO, GEO, and
Highly Elliptical Orbits (HEO), flying nodes such as UAVs, and HAPs. In the decade
2020–2030, it is envisioned to deploy LEO and mega-LEO constellations (with the
number of satellites on the order of several thousands) to provide global Internet
services. However, to achieve the vision of satellite-6G network integration, appropriate
Reference Architectures (RAs) need to be designed and developed.
In [4], RA-3 focuses on 6G-satellite integration where the satellite acts as an access
network. In this work, we discussed a unique application over RA-3, SBHS [5], which
emphasizes the impact of satellite broadband Internet on rural and remote areas around
 Special Issue of IEEE
Advanced Solutions for 6G satellite
Systems Workshop

the world with the help of the next generation mega satellite constellations. The SBHS
focuses on achieving the goal of connecting the digitally unconnected population. SBHS
deals with hosting an entire content-server in the LEO satellites to reduce the end-to-end
latency. In this work, the RA-3 is utilized to deliver the SBHS as an application.
The remaining of the paper is organized as follows: Section II introduces the reference
satellite architecture for 6G satellite network integration. Section III details possible use
cases and case studies using RA-3. Section IV explains the simulation environment used
to study the SBHS approach using RA-3. Section V provides performance results related
to SBHS. Finally, Section VI concludes the paper.

2. REFERENCE ARCHITECTURES TOWARD 6G-SATELLITE

NETWORK INTEGRATION
One of the most significant aspects of 6G-satellite network integration is the reference
architecture. In this paper, we consider three RAs: (i) non-virtualized satellite networks
(RA-1), (ii) separate virtualized satellite networks (RA-2), and (iii) integrated virtualized
6G-satellite networks (RA-3) [4]. The interfaces defined for RAs are closely related links
to those specified in 3GPP TR 38.821 Rel. 16 [6].
For the 5th Generation (5G) communication systems, two RAs were outlined in [7] for 5G-
satellite integration. In the first RA (RA-1), named as non-virtualized 5G-satellite
networks, the satellite infrastructure is envisioned to use as the backhaul infrastructure
for 5G-terrestrial networks. In the second RA (RA-2) [7], named as separately virtualized
5G-satellite networks, satellite network forms a virtualized network infrastructure,
however, separated from the terrestrial 5G virtualized network.
The RA for 6G-satellite network integration is an extended version of RA-2, named
integrated virtualized 6G-satellite networks (RA-3) [4]. The scheme RA-3 is expected to
include only a single Management and Network Orchestration (MANO) for the satellite
and terrestrial segments. In RA-3, the control plane design, control operations, network
slicing, network functions, network interfaces, network orchestration approaches, and
functional aspects of the network follow the same pattern for both the virtualized satellite
and terrestrial 6G networks. The virtualized network (i.e., integrated satellite and
terrestrial 6G) maintains a single MANO module for its respective management activities.
The responsibilities to carry out resource management, routing of packets, channel
management, slice management, edge computing decisions, and federation functions
rest with the integrated MANO module of the network.
Besides the functions outlined in 3GPP, two additional network functions for RA-2 and
RA-3 are defined: Satellite Edge-computing Function (SEF) and Satellite Network
Federation Function (SNF). A part of the SEF is already considered in [8] with a multi-
tier caching framework for Satellite Edge Computing (SEC) for 5G satellite integration
with inter-satellite cache transfer.

3. REFERENCE ARCHITECTURE USE-CASES

A set of 13 use-cases were proposed in [7] for 5G terrestrial networks employing satellite
network infrastructure as the backhaul infrastructure. The use-cases include
communication between Distributed Unit (DU) and Centralized Unit (CU) through single-
 Special Issue of IEEE
Advanced Solutions for 6G satellite
Systems Workshop

hop/multihop links using federated/non-federated LEO network. Further, the use-cases

with UAVs and HAPs, employed for backhaul infrastructure, were discussed in [7]. HAPs
can be served as flying airborne base stations to connect the unconnected or the under-
connected [9]. HAPs can be deployed up to an altitude of 30 km [10] and offer round-trip
latency of the order of less than 1.2 ms. In fact, HAPs can be integrated to the backbone
network. HAPs act as a tower-in-the-air by relaying data between the BS and
eNodeB/eNodeG where either mobile UEs or access points are located in under-served
regions [11].
The backhaul service scenarios for NTN are described in [4] where the satellite network
infrastructure is used for direct access. The new use-cases for direct access to satellite
networks focus on the entire spectrum of NTNs including HAPs, UAVs, and satellites.
Two major directions are considered with an eNodeB/gNodeB being (i) on-NTN or (ii)
on-ground. Further, three direct access modes for NTN nodes are discussed: (i)
gNodeB/eNodeB onboard the NTN node (satellite/HAPs), (ii) relay access where the
backhaul traffic is forwarded to another high-level node by the NTN node, and (iii) bent-
pipe by NTN node (satellite/HAPs). The entire functions of eNodeB/gNodeB are included
in the assumption of eNodeB/gNodeB irrespective of its location in NTN. The use-cases
are divided according to LAPs, HAPs, and satellites. Using LAPs and HAPs as
gNodeB/eNodeBs, 16 use-cases are described where seven use-cases are defined for
satellites acting as gNodeB/eNodeB [4].
The physical layer assumptions for the use-cases discussed in [12] include: (i) the use
of LEO satellites constellation, (ii) circular orbit around the Earth with a typical beam
footprint of 100–1000 km, (iii) transparent or regenerative LEO satellite payload, (iv) no
inter-satellite links, (v) fixed or movable beams (moving or fixed footprint on the ground),
(vi) below 6 GHz frequency band, and (vii) direct or indirect access network (indirect
access via terrestrial gateway or HAPs).

a. Space-Based Hosting Service

In [5], an approach for SBHS was proposed to deploy content servers in LEO and GEO
satellites. The SBHS framework was envisioned to play an important role in
accomplishing the goal of connecting the digitally unconnected population. In [5], the
SBHS demonstrated an example of hosting a complete content server in a LEO satellite.
Iridium-NEXT satellite constellation [13] was used for the simulation studies. The
simulation studies showed the feasibility of hosting an entire content server in space to
achieve ultra-low latency compared to the traditional satellite-based web services.

Figure 1 illustrates the use of RA-3 in SBHS. SBHS uses the regenerative satellite
payload where an entire content server is present. LEO satellites are used to host the
content server in SBHS. To reduce the delay, the LEO satellite serves directly to the
users. The integrated 6G-satellite control plane is located in LEO satellites. The
terrestrial 6G part is not present as SBHS only uses LEO satellites to serve the end
users. Further, the networking functions present in the MANO are modified to suit the
SBHS.

4. SIMULATION SETUP
 Special Issue of IEEE
Advanced Solutions for 6G satellite
Systems Workshop

We considered Iridium-NEXT satellite constellation for our simulation. A Python-based

library, skyfield [14] is used to create the LEO satellite network topology. The network
parameters used in the simulation are described in TABLE I. Iridium-NEXT constellation
consists of 75 satellites placed in polar orbits of altitude 780 km and an inclination of
86.4o. We assumed the users are present in a 10 × 10 km2 area within India. Multiple
user densities were considered in the simulation. The uplink and downlink end-to-end
data rates between UE and LEO satellites are set to 100 Mbps and 200 Mbps,
respectively. The CPU processor capacity in LEO satellites is assumed to be 1 GHz. The
size of a request message from user to satellite is taken from a uniform distribution
between 1–300 Bytes. The average content size is assumed as 5 KB which is present in
the LEO satellite. The simulation experiment is carried out for 100 minutes which is the
average orbital period of satellites in the Iridium-NEXT satellite constellation.

5. PERFORMANCE ANALYSIS
We evaluated the performance of the proposed SBHS through average end-to-end
delay. Figure 2a illustrates the average end-to-end delay, round trip propagation delay,
and queuing delay. We observed the lowest average end-to-end delay of 7.75 ms. In
Figure 2a, the significant contributing factors to the end-to-end delay are round trip
propagation delay and queuing delay. We considered two queues in our simulation. One
queue is used to collect the user requests and the second queue is used to transmit the
content to the user. The propagation delay contributions to the end-to-end delays for
2000, 5000, and 10,000 users are 91%, 91.6%, and 88%, respectively. Similarly, queuing
delay contributions to the same user densities are, respectively, 26.5%, 30%, and 33.5%.
In 5G systems, the delay related Key Performance Indicator (KPI) is 1 ms. However, it
is impossible to achieve a 1 ms delay with Iridium-NEXT satellite constellation. This is
due to the high propagation delay which depends on the altitude of content server
satellites. Therefore, the propagation delay is not reducible. The queuing delay can be
reduced by increasing (i) the number of cores and frequency of the processor and (ii) the
number of transmission channels. However, reducing queuing delay is not sufficient as
the major contributing factor to the end-to-end delay is propagation delay.
Figure 2b illustrates the other minor delays that contributed to the end-to-end delay. The
minor delays consist of uplink transmission delay, downlink transmission delay, delay to
search the contents in the database, and delay for retrieving contents from the database.
The minor delays are in the microseconds range. Therefore, the significance of those
delay components is less. The end-to-end delay experienced by different densities of
users is similar. This is due to the small content size.

6. Conclusion
In this paper, we considered the reference architectures based on both backhaul and
direct access of NTN elements for 6G systems integration with the satellite networks. An
integrated virtualized 6G-satellite architecture was discussed based on NTN elements
such as LEO, MEO, GEO, HEO satellites, UAVs, and HAPs. Finally, an application
services case study of Space Based Hosting Service (SBHS) was presented. We
observed a minimum average end-to-end delay of 7.75 ms for the country India using
SBHS approach.
Acknowledgements
 Special Issue of IEEE
Advanced Solutions for 6G satellite
Systems Workshop

This work is funded in parts by IIT Palakkad Technology IHub Foundation

Doctoral Fellowship IPTIF/HRD/DF/032.

Table 1. SIMULATION PARAMETERS

Parameter Value
User population 2000, 5000, and
10,000
Number of satellites 75
Request message size 1 -- 300 bytes
Processor frequency 1 GHz
End-to-end uplink data 100 Mbps
rate
End-to-end downlink 200 Mbps
data rate
CPU cycle/bit 10
Queue length 100
Simulation time 100 minutes
 Special Issue of IEEE
Advanced Solutions for 6G satellite
Systems Workshop

Fig 1: SBHS over Reference Architecture-3

References
1. International Telecommunication Union World Telecommunication/ICT indicators database,
“Individuals using the Internet (percentage of population),” 2020. [Online]. Available:
https://data.worldbank.org/indicator/IT.NET.USER.ZS (accessed:02-25-2021)
2. International Telecommunication Union (ITU) publications, “Measuring digital development facts
and figures 2019,” 2020. [Online]. Available:
https://www.itu.int/en/ITUD/Statistics/Documents/facts/FactsFigures2019.pdf (accessed:02-25-
2021)
3. International Telecommunication Union (ITU) publications, “Measuring digital development facts
and figures 2021,” 2020. [Online]. Available:
 Special Issue of IEEE
Advanced Solutions for 6G satellite
Systems Workshop

https://www.itu.int/en/ITUD/Statistics/Documents/facts/FactsFigures202.pdf (accessed:09-19-
2022)
4. IEEE Future Network Satellite Working Group Team, “International network generations roadmap
(INGR)-2022 edition (satellite),” 2022. [Online]. Available:
https://futurenetworks.ieee.org/images/files/pdf/INGR-2022-Edition/IEEE INGR Satellite Chapter
2022-Edition-FINAL.pdf (accessed:10-25-2022)
5. D. Dalai, S. Babu, V. B.S., and B. S. Manoj, “A novel space based hosting approach for ultra low
latency web services,” Feb. 2022. [Online]. Available:
https://www.techrxiv.org/articles/preprint/A_Novel_Space_Based_Hosting_Approach_for_Ultra_L
ow_Latency_Web_Services/18666455
6. 3GPP TR 38.821, “Solutions for nr to support non-terrestrial networks (NTN),” (Release 16)
7. IEEE Future Network Satellite Working Group Team, “International network generations roadmap
(INGR)-2021 edition (satellite),” 2021. [Online]. Available:
https://futurenetworks.ieee.org/images/files/pdf/INGR_2021_Edition/IEEE_INGR_Satellite_2021E
d.pdf (accessed:10-25-2022)
8. D. Dalai, S. Babu, and B. S. Manoj, “On using edge servers in 5G satellite networks,” in 2020 IEEE
3rd 5G World Forum (5GWF), Oct. 2020, pp. 553–558.
9. E. Yaacoub and M.-S. Alouini, “A key 6G challenge and opportunity—connecting the base of the
pyramid: A survey on rural connectivity,” Proceedings of the IEEE, vol. 108, no. 4, pp. 533–582,
Mar. 2020.
10. A. Alsharoa and M.-S. Alouini, “Improvement of the global connectivity using integrated satellite-
airborne-terrestrial networks with resource optimization,” IEEE Transactions on Wireless
Communications, vol. 19, no. 8, pp. 5088–5100, Apr. 2020.
11. E. Yaacoub and M.-S. Alouini, “Efficient fronthaul and backhaul connectivity for IoT traffic in rural
areas,” IEEE Internet of Things Magazine, vol. 4, no. 1, pp. 60–66, Dec. 2021.
12. 3GPP TR 36.763, “Study on narrow-band Internet of Things (NBIoT)/enhanced machine type
communication (eMTC) support for non terrestrial networks (NTN),” (Release 17).
13. “IRIDIUM satellite communication,” 2022. [Online]. Available: https://www.iridium.com/
(accessed:02-25-2022)
14. B. Rhodes, “Skyfield: High precision research-grade positions for planets and Earth satellites
generator,” Astrophysics Source Code Library, record ascl:1907.024, p. ascl:1907.024, Jul. 2019.
 Special Issue of IEEE
Advanced Solutions for 6G satellite
Systems Workshop

Debabrata Dalai (Graduate Student Member, IEEE) received the bachelor’s

degree in electronics and telecommunication engineering from Biju Patnaik
University of Technology (BPUT), Rourkela, India, and the master’s degree in
computer science and engineering (Advanced Networks) from the Indian
Institute of Information Technology and Management, Gwalior, India. He is
currently a Senior Research Fellow with the Department of Avionics, Indian
Institute of Space Science and Technology, Thiruvananthapuram, India. He
serves as the current chair of IEEE IIST Student Branch. His research interests
include advanced wireless networks, satellite-IoT networks, satellite edge
computing, and 5G/6G satellite integration.
Sarath Babu (Member, IEEE) received his Ph.D. degree from Indian Institute
of Space Science and Technology, Thiruvananthapuram, India in 2021. He
completed his master’s degree in Computer Science and Engineering
(Information Security) from National Institute of Technology, Calicut, India, and
bachelor’s degree in Information Technology from Mahatma Gandhi
University, Kottayam, India. He is currently working as a Research Scientist at
the Department of Electrical and Computer Engineering, Iowa State University,
USA, and affiliated to the Center for Wireless, Communities, and Innovation
(WiCI). His primary focus is on the project ARA-Wireless Living Lab for Smart
and Connected Rural Communities, a part of NSF Platforms for Advance
Wireless Research program. His research interests include next generation
wireless infrastructures, software defined networks, complex networks,
systems security, and intelligent transportation systems.
B. S. Manoj (Senior Member, IEEE) received the Ph.D. degree in computer
science and engineering from the Indian Institute of Technology Madras in
2004. He was an Assistant Research Scientist and a Lecturer with the
Electrical and Computer Engineering Department, University of California at
San Diego. He is currently a Professor with the Department of Avionics, Indian
Institute of Space Science and Technology, Thiruvananthapuram, India. His
current research interests include complex networks, ad hoc wireless
networks, delay tolerant networks, next generation wireless architectures,
wireless sensor networks, wireless mesh networks, satellite networks, 6G-
satellite integration, and quantum computing.