Available via license: CC BY 4.0
Content may be subject to copyright.
1
Point-to-Point Communication in Integrated
Satellite-Aerial Networks: State-of-the-art
and Future Challenges
Nasir Saeed, Senior Member, IEEE, Heba Almorad, Student Member, IEEE, Hayssam Dahrouj, Senior
Member, IEEE, Tareq Y. Al-Naffouri, Senior Member, IEEE, Jeff S. Shamma, Fellow, IEEE, and Mohamed-Slim
Alouini, Fellow, IEEE
Abstract—This paper overviews point-to-point (P2P) links for
integrated satellite-aerial networks, which are envisioned to be
among the key enablers of the sixth-generation (6G) of wireless
networks vision. The paper first outlines the unique charac-
teristics of such integrated large-scale complex networks, often
denoted by spatial networks, and focuses on two particular space-
air infrastructures, namely, satellites networks and high-altitude
platforms (HAPs). The paper then classifies the connecting P2P
communications links as satellite-to-satellite links at the same
layer (SSLL), satellite-to-satellite links at different layers (SSLD),
and HAP-to-HAP links (HHL). The paper overviews each layer
of such spatial networks separately, and highlights the possible
natures of the connecting links (i.e., radio-frequency or free-space
optics) with a dedicated overview to the existing link-budget
results. The paper, afterwards, presents the prospective merit
of realizing such an integrated satellite-HAP network towards
providing broadband services in under-served and remote areas.
Finally, the paper sheds light on several future research directions
in the context of spatial networks, namely large-scale network
optimization, intelligent offloading, smart platforms, energy effi-
ciency, multiple access schemes, and distributed spatial networks.
Index Terms—Integrated satellite-aerial networks, spatial net-
works, satellites, high-altitude platforms, broadband services.
I. INTRODUCTION
Connectivity is the backbone of modern digital economy
with over three billion people connected worldwide, and more
than 14 billion devices connected through the Internet core
network. Although the wireless coverage has spread substan-
tially over the past two decades, almost half of the world’s
population remains unconnected [1]. With the data deluge in
terms of global services and user-equipments, the number of
connected devices is expected to surpass 50 billions, which
poses stringent burdens on the current telecommunications
terrestrial infrastructure [1]. Therefore, developing novel con-
nectivity solutions to fulfill such enormous demands becomes
an indispensable necessity.
Nasir Saeed, Tareq Y. Al-Naffouri, Jeff S. Shamma and Mohamed-Slim
Alouini are with the Department of Computer, Electrical and Mathematical
Sciences and Engineering (CEMSE), King Abdullah University of Science
and Technology (KAUST), Thuwal, Makkah Province, Kingdom of Saudi
Arabia, 23955-6900.
Heba Almorad is with the Department of Electrical and Computer Engineer-
ing, Effat University, Jeddah 22332, Saudi Arabia.
Hayssam Dahrouj and Jeff S. Shamma are with the Center of Excellence
for NEOM Research, King Abdullah University of Science and Technology,
Thuwal 23955-6900, Saudi Arabia.
A recent trend for boosting ground-level communication is
by enabling connectivity from the sky as a means to connect
the unconnected and super-connect the already connected,
a theme that falls at the intersection of the ongoing sixth-
generation (6G) wireless networks initiatives [2]–[4]. Towards
this direction, integrated satellite-aerial networks, also known
as spatial networks (SNs), have emerged as essential enablers
for serving remote areas and enhancing the capacity of the
existing wireless systems [2]–[6]. Thanks to their capabil-
ities at connecting wireless platforms of different altitudes,
SNs provide high data rates for terrestrial wireless backhaul
networks [7], and enable global Internet services [8]. While
the original focus of SNs is mainly on satellites deployment,
recent SNs studies include other non-terrestrial networks that
operate at a comparatively lower altitude, i.e., communications
infrastructures at the stratosphere and troposphere layers [9].
Besides connectivity, SNs have plenty of valuable applications,
e.g., surveillance, weather forecasting, earth observation, nav-
igation, and climate monitoring [10]–[12].
Spatial networks consist of a plurality of nodes (also called
spatial elements) in two- and three-dimensional spaces, which
form single and multilayer architectures. Such nodes can be
satellites, high-altitude platforms (HAPs), tethered balloons,
or unmanned aerial vehicles (UAVs) [13]. The type of ar-
chitecture then depends on the altitude of nodes. While the
nodes at the same altitude are called single-layer nodes, the
nodes at different altitudes are called multilayer nodes. The
multilayered architecture often offers more degrees of freedom
than the single-layer, and can provide a global connectivity
solution since the multilayered architecture combines several
layers, and exploits the compound benefits of the different
layers at the different altitudes [14]. Fig. 1 illustrates a generic
multilayered architecture of SNs where each layer is at a
different altitude from the Earth’s surface, i.e., deep space
(>35,838 km), geo-synchronous Earth orbit (GEO) (12000-
35,838 km), medium Earth orbit (MEO) (2000-12000 km),
low Earth orbit (LEO) (200-2000 km), stratospheric (17-22
km), and aeronautical (0.15-17 km) [15]. The spatial elements
in each layer can relay data in a multihop fashion among the
different nodes of SNs, thus converting a long-range single-
hop link into short-range multi-hop links, thereby reducing the
overall propagation delay and improving the overall data rate
[16].
The multi-hop links can be established within a single
arXiv:2012.06182v1 [eess.SP] 11 Dec 2020
2
Fig. 1: Illustration of a multilayered SN with satellites, HAPs, and UAVs.
layer (intra-layer) of SNs or between nodes of two or more
different layers (inter-layer), as illustrated in Fig. 1. One can
then categorize the SNs communications links as satellite-to-
satellite links at the same layer (SSLL), satellite-to-satellite
links at different layers (SSLD), HAP-to-HAP links (HHL),
and UAV-to-UAV links (UUL), respectively. Satellites, HAPs,
and LAPs are equipped with on-board processing (OBP) ca-
pabilities to establish such links, allowing the communication
between different elements on the same layer or even at
different layers in SNs [17]. One significant difference between
the terrestrial networks and SNs is that the latter consists of
network topologies with significantly heterogeneous network
nodes within the well-spread space-air layers, as illustrated
in Fig. 1. The links in such a multilayer network can be
established using both radio-frequency (RF) waves and free-
space optics (FSO), as discussed in details later in the paper.
In the current practice, radio frequencies in the microwave
band are used to establish point-to-point (P2P) wireless links
among the different entities of SNs. For example, the common
data link (CDL) that is designed by the U.S Department of
Defense uses Ku (12-18 GHz) and Ka (26-40 GHz) frequency
bands to transmit data for long P2P communication between
HAPs and terrestrial stations [18]. However, CDL’s limited
spectrum constraints limit its data rate between 274 Mbps
to 3 Gbps, which do not satisfy the demand for high-speed
wireless links [6] [18]. In this context, U.S. Defense Ad-
vanced Research Projects Agency (DARPA) started a program
called “Free-space Optical Experimental Network Experiment
(FOENEX)” to develop links that can transmit data using FSO
at a much higher speed. In 2012, FOENEX successfully estab-
lished the first FSO link to allow a 10 Gbps transmission rate
for airborne platforms. After further improvement, it turned
out that FSO can provide up to 100 Gbps P2P links using
wavelength-division multiplexing (WDM), which is superior
than the average rates of RF-based systems [19]. FSO tech-
nology is also energy-efficient, secure, and license-free, which
make it a strong candidate for space-borne P2P communication
deployment [20] [21]. FSO technology is, however, generally
vulnerable to the environment and cannot operate efficiently
in a rainy, snowy, or foggy weather. Also, the FSO links
require perfect alignment between the transmitter and receiver
of the moving platforms [22], which is often handled using
a variety of alternative techniques [23]–[25]. Consequently,
DARPA launched another program to investigate ways of
establishing the same 100 Gbps with all-weather tolerance
capability. Towards this direction, the program investigated the
mmWave spectrum (30-300 GHz) and exploited high-order
modulation and spatial multiplexing techniques to attain the
desired data rate for a range of 200 km intra-layer link, and
100 km for the inter-layer link in the stratospheric region [26].
DARPA then identified mmWave technology as the suitable
solution for airborne communication. The results showed
an outstanding performance achieving 100 Gbps under the
atmospheric attenuation, and cumulus loss with less than 0.3
dB/km in the E-band (71–76 GHz and 81–86-GHz).
Other interesting ongoing projects on SNs P2P links adopt
hybrid RF/FSO [27], as a means to combine the mutual
advantages of both RF and FSO. Such systems operate by
3
switching to low-capacity RF links in bad weather conditions,
or to high-capacity FSO links under perfect transceivers align-
ment and suitable weather conditions. One such hybrid project
is Integrated Aerial Communications (FaRIA-C) headed by
DARPA [28]. This project started in 2019 to develop simul-
taneous hybrid links that switch between FSO and RF, based
on the environment suitability. In other words, whenever the
weather obscures the Line-of-Sight (LoS), the system switches
from FSO to RF. FaRIA-C achieves up to 10 Gbps link
capacity when operating at FSO and 2 Gbps at RF band [28].
Despite their promising capabilities, hybrid FSO/RF systems
still face various challenges, such as scheduling, scalability
of the network, and quality of service (QoS) constraints, as
highlighted in [29]. In Table I, we summarize some of the well-
known projects that use different communication technologies
for enabling P2P links in SNs.
A. Related Review Articles
Due to the significance of P2P communications in SNs,
there is a plethora of review articles, each discussing different
aspects of SNs [4], [15], [20], [37]–[49]. For instance, ref-
erence [40] reviews UAVs-based ad hoc networks, including
the application scenarios, design characteristics and consid-
erations, communication protocols, and open research issues.
Chen et al. provide a survey focusing on the coverage problem
in UAV networks until 2014 [41]. Then, reference [42] further
extends the literature on UAV communication and coverage is-
sues such as routing, seamless handover, and energy efficiency
until 2016. [47] presents an updated UAV communications
survey that discusses the practical aspects, standardization ad-
vancements, and security challenges. Furthermore, the authors
in [47] enumerate the 3GPP study items for maximizing the
UAV opportunities in 4G and 5G applications. Moreover, [45]
surveys channel modeling for UAV communications, including
channel characterization, channel modeling approaches, and
future research directions.
From the stratospheric layer perspective, reference [37]
explores various facets of P2P wireless communications in
HAPs, including channel modeling, interference, antennas, and
coding. The study in [38] is further narrowed down to FSO for
P2P wireless links in HAPs, mainly focusing on acquisition,
tracking, and pointing (ATP) issues. Recently, the authors in
[48] present a comprehensive and up-to-date survey on how to
extend coverage and resolve capacity issues in rural areas us-
ing HAPs. The focus in [48] is on HAPs regulations, projects,
network topologies, and handover mechanisms. Moreover, the
authors in [15] conduct extensive research on heterogeneous
SNs, i.e., HAPs and LAPs, but does not come across the
satellites aspects of SNs.
Reference [20] presents more detailed insights on SNs, such
as ATP for space-based optical links, hybrid RF/FSO solution,
MIMO, and adaptive optics. Unlike the above articles, the
review [20] addresses all the layers of SNs; however, it
focuses mainly on the satellites layer by discussing various
satellite system aspects, medium access protocols, networking,
testbeds, air interface, and future challenges [49].
In terms of space networks, Mukherjee et al. survey the
communication technologies and architectures for satellite
networks and interplanetary Internet, demonstrating the notion
of delay-tolerant networking (DTN) for deep space networks
[39]. Furthermore, Krishnan et al. present an extensive study
on diverse inter-satellite link design issues based on the last
three layers of the open system interconnection (OSI) [43].
[43] proposes employing DTN protocols as a solution to
the problems surveyed, detailing the required design param-
eters for inter-satellite communications. Moreover, dynamic
resource allocation algorithms and schemes in integrated GEO
satellite-ground networks are reviewed in [46]. [4] highlights
various issues for small satellites called CubeSats, discussing
the coverage, different constellation designs, upper layer is-
sues, and future research challenges. Moreover, [20] and
[44] present a study on FSO communications for satellites,
including uplinks, downlinks, and ISL links. In Table II, we
summarize the contributions of related review articles.
B. Contributions of our Paper
Unlike the above-mentioned surveys which only focus on
a single non-terrestrial network layer, i.e., either satellites
or HAPs, our current paper focuses on P2P links for a
multi-layered spatial network. The main motivation of this
survey originates from their importance of studying the unique
characteristics of spatial networks and the P2P interconnecting
links in light of 6G large-scale complex networks. To this
end, the paper presents the studies on wireless communication
technologies for each layer separately, including satellites and
HAPs layers. In conjunction, the paper overviews two possible
alternatives for intra- and inter-satellite links, mainly FSO
and RF connections, and discusses various possibilities for
enabling P2P links among HAPs and from HAPs to the
ground station. To best illustrate the compound benefits of
the different layers integration, the paper then sheds light on
the integrated satellite-HAP network as a means to provide
broadband services in underserved areas. Finally, the paper
presents several future research directions in the context of
spatial networks, including large-scale network optimization,
intelligent offloading, smart platforms, energy efficiency, mul-
tiple access schemes, and distributed spatial networks.
C. Paper Organization
The rest of the paper is organized as follows. Section
II presents P2P links in satellite networks, covering both
intra- and inter-layer links. Moreover, it provides link budget
calculation for both RF and FSO-based inter-satellite links.
We report the studies on P2P links in HAP-based networks
in Section III, discussing both inter-HAP links and HAPs-
to-ground communication. Section IV provides a review of
integrated satellite-HAP networks to improve the reliability,
coverage, and scalability for future 6G wireless communica-
tion systems. We present numerous future research directions
in Section V, and then we conclude the paper in Section IV.
II. P2P LINKS IN SATELLITE NETWOR KS
With the emergence of the new space economy, satellite
communication is getting more advanced in providing the
4
TABLE I: List of a few projects that uses P2P wireless communication links.
Project Technology Platform Link type Data rate Distance
(km)
Year
Iridium [30] RF (L-band) Satellites LEO-to-LEO 25 Mbps - 1997
SILEX [31] FSO (847nm-
819nm)
Satellites GEO-LEO Link 50 Mbps 45000 2001
IRON-T2 [29] FSO (1556.1nm)
/ RF (X/Ku-band)
- LAP-to-LAP 2.5-40/0.274 Gbps 50-200 2007
FALCON [32] FSO Aircrafts LAP-to-LAP 2.5 Gbps 130 2010
LAC [33] FSO (532 nm) Airships,
UAVs
HAP-to-HAP, LAP-to-
LAP, and HAP-to-LAP
10-40 Gbps 200 2014
CURfEGC [34] RF (UHF) Satellites MEO-to-MEO 1-2 Mbps 31,400 2016
QB50 [35] RF (VHF/UHF) Satellites LEO-to-LEO 0.5-10 kbps 90 2017
Stellar [36] FSO (915nm) Satellites LEO-to-LEO 100 Mbps 1000 2020
Fig. 2: Illustration of different satellite topologies with satellite-to-satellite links at same layer.
Internet from space. The satellite networks consist of many
satellites at different altitudes, revolving in various types of
constellations, using different frequency bands with distinct
coverage. Therefore, it is critical for the satellite networks to
take into account the essential characteristics, such as altitude,
constellation, and operating frequency band, to achieve a
specific goal. For example, the higher the satellite is, the
wider the area it covers (a GEO satellite can cover around
30%of the Earth’s surface, while a group of MEO and LEO
satellites is required to cover the same area). On the other
hand, MEO and LEO satellites provide shorter paths than
GEO, resulting in less propagation delay. Also, satellites in low
altitude constellations move faster, leading to a higher Doppler
effect. Besides, the GEO, MEO, and LEO, constellations can
be designed in such a way to increase the dwell time in certain
parts of the world, for example, in highly elliptical orbits
(HEO) [49].
Apart from the constellation design, enabling P2P links
among the satellites is crucial for relaying the data. There are
two possible relaying methods in satellite networks, namely
amplify-and-forward (AF) and decode-and-forward (DF) [50].
Satellites that use AF techniques are known as transparent
satellites because they only amplify the received signal and
forward it to the neighboring satellites or the ground station.
On the other hand, DF satellites, or regenerative satellites,
decode the incoming signal and perform signal processing to
mitigate the interference and regenerate it. Besides relaying,
the selection of a routing topology is critical for efficient com-
munication between the satellites and the ground segments,
or between the satellites. Typically, there are three topologies
(i.e., star, mesh, and line) used in satellite networks based
on the target application [49]. As depicted in Fig. 2, in a
star topology, satellites are connected to a central node that
controls their interconnections. In contrast, in a mesh setup,
all satellites are directly connected [51]. Moreover, in line
topology, the satellites are communicating with their neighbors
only, following a line structure, as shown in Fig 2. Among
these topologies, the star is by far the most popular for master-
slave networks since it reduces the chances of network failures.
However, mesh topology has more degree of freedom and less
latency at the cost of more complexity because it enables more
SSLL. Apart from the topologies, it is crucial to analyze the
link design for both RF and optical-based SSLL to ensure
sufficient connectivity and cooperation between the satellites.
A. Satellite-to-Satellite Links at Same Layer (SSLL)
Scientists from NASA, ESA, and DARPA studied both
intra- and inter-layer P2P satellite links, for over a decade.
A.C. Clarke introduced the concept of satellite-to-satellite
links in 1945 [52]. Afterwards, SSLL became commonly used
in satellite networks to offer cost-effective communication
services. In contrast to the satellite-to-ground link, which is
5
TABLE II: Comparison of this paper with the existing surveys.
Ref. Platform
Type
Area of Focus Year
Karapantazis et al.
[37]
HAPs Presents possible architectures of HAPs, system structure, channel modeling,
antennas, coding techniques, resource allocation techniques, and applications.
2005
Fidler et al. [38] HAPs Outlines FSO communication technology, system design requirements, data
transmission and correction techniques, and experimental field trials for
HAPs.
2010
Mukherjee et al. [39] Satellites Discusses architectures, communication technologies, networking protocols,
interplanetary Internet, and open research challenges for satellite networks.
2013
Bekmezci et al. [40] LAPs Focuses on design characteristics, routing protocols, applications and open
research issues for UAV networks.
2013
Chen et al. [41] LAPs Discusses the coverage issues for UAV networks. 2014
Gupta et al. [42] LAPs Reviews major issues in UAV communication networks, including mobility,
limited energy, and networking.
2016
Krishnan et al. [43] Satellites Presents various design parameters based on the last three OSI model for
small satellite networks, such as modulation-and-coding, link design, antenna
type, and different MAC protocols.
2016
Kaushal et al. [20] Satellites Discusses space-based optical backhaul links and their applications. 2017
Son et al. [44] Satellites Highlights the importantence of FSO communications for inter-satellite links. 2017
Khuwaja et al. [45] LAPs Surveys channel modeling techniques for UAV communications. 2018
Peng et al. [46] Satellites Presents dynamic resource allocation schemes for integrated satellite and
terrestrial networks.
2018
Cao et al. [15] HAPs/
LAPs
Discusses design parameters and protocols for HAPs and LAPs communica-
tion networks.
2018
Fotouhi et al. [47] LAPs Surveys practical aspects, standardization, regulation, and security challenges
for UAVs-based cellular communications.
2019
Arum et al. [48] HAPs Focuses on coverage and capacity issues using HAPs. 2020
Saeed et al. [4] Satellites Presents channel modeling, modulation-and-coding, coverage, constellation
issues, networking, and future research directions for CubeSat communica-
tions.
2020
Kodheli et al. [49] Satellites Discusses the recent technical advances in scientific, industrial, and standard-
ization for satellite communications.
2020
This paper SNs Outlines the unique characteristics of large-scale complex SNs, surveys
various wireless communication technologies to implement P2P links’ in SNs,
and points out several promising research directions.
2020
a duplex link, SSLL are mainly simplex links where the path
length is measured by the LoS distance between any two
satellites [53]. In current systems, SSLL can be established by
using either RF or FSO technologies [54]. In the following,
we discuss the link budget analysis for both RF and optical
SSLL.
1) RF Link Budget: In satellite communications, RF SSLL
are the most widely used communication links because of their
reliability and flexible implementation. Before calculating the
link budget, it is essential to know the functional modulation
and coding schemes used in RF-based links. Mainly, coherent
systems such as Binary Phase Shift Keying (BPSK) are more
desirable due to their lower power requirements to achieve
a given throughput and bit error rate (BER). Nevertheless,
the coherence capability produces delays as it takes time to
lock the transmitted signal in the receiver terminal. Unlike
coherent systems, non-coherent systems such as Frequency
Shift Keying (FSK) require more transmitting power to achieve
the same throughput and BER with less delay. Another popular
modulation scheme for RF-based SSLL is Quadrature Phase
Shift Keying (QPSK), which provides twice the bandwidth
than a typical BPSK. QPSK, however, suffers from phase
distortion because of the channel values, leading to system
degradation, which is often solved using differential PSK in
order to improve the overall spectral efficiency through striking
a trade-off between power requirements and spectral efficiency
[43].
For a given modulation scheme and under a non-coding
assumption, the parameters used in calculating the link budget
for RF-based SSLL can be described as a function of the
satellite transmit power (Pt), the distance between satellites
(d), achievable data rate (Rb), operating wavelength (λ), and
diameter of the transmit antenna’s aperture (D). For simplicity,
the radiation of the transmitting antenna is assumed to be
isotropic, where the radiation intensity is the same in all
directions. Therefore, the gain of the transmitter and receiver
6
antennas Gtand Grcan be calculated as follows:
Gt=Gr=4πA
λ2,(1)
where A=πD2
4is the aperture of the antenna. Besides the
gain of the transmitter and receiver antennas, path loss Lpis
critical in the analysis and design of SSLL. Such pathloss can
be calculated at the receiver antenna as follows
Lp=4πd
λ2
,(2)
Based on the path loss, the received power is calculated as,
Pr=PtGtGr
Lp
.(3)
To determine whether the received power is sufficient to
establish a satellite-to-satellite link or not, we need to find
the required signal-to-noise-ratio (SNR), assuming that the
noise is additive white Gaussian noise (AWGN). Such noise
mimics the random processes effect in space, where the only
communication impairment is the white noise. Besides, the
required SNR primarily depends on the used modulation
scheme and the target bit error probability (Pb) [24]. For
instance, if the modulation scheme is BPSK, then the SNR
required to achieve Pbfor the RF-based SSLL can be written
as
γreq =Eb
No
=Pr
kT RbB,(4)
where Eb
Nois the bit-energy per noise spectral-density, Bis
the bandwidth in Hertz, k= 1.38 ×10−23 is the Boltzmann
constant, and T= 300K is the absolute temperature [55].
Hence, Pbis calculated as
Pb=1
2erfc(√γreq ),(5)
where erfc(·)is the complimentary error function.
TABLE III: Parameters for RF-based link budget calculation
Parameter Value
Transmitted power Pt2 W
Satellite antenna gain Gt&Gr0 dBi
Data rate Rb1 Mbps
Bandwidth B0.5 MHz
Antenna aperture area A7.84 cm2
Absolute temperature T300 K
We next give some numerical insights that highlight the
above link-budget characterization. Consider RF-based SSLL
among satellites orbiting in LEO. We first analyze the impact
of distance and operating frequency on the received power
and SNR. By varying the distance between satellites and the
operating frequency of their interconnections, the received
power is then calculated based on the above equations. Table
III summarizes the parameters for calculating the RF-based
link budget. From Fig. 3, we observe that the received power
is inversely proportional to the distance between satellites and
the frequency. At the same distance, SSLL operating at a
0 10 20 30 40 50 60 70 80 90 100
-140
-130
-120
-110
-100
-90
-80
Fig. 3: The received power for RF-based SSLL with varying
link distances.
0 10 20 30 40 50 60 70 80 90 100
-50
-40
-30
-20
-10
0
Fig. 4: The energy-per-bit to noise spectral density for
RF-based SSLL with varying link distances.
lower frequency results in a higher received power. This is
mainly due to the frequency-dependent path loss, i.e., since
the path loss increases at higher frequencies, the level of the
received power decreases. On the basis of the International
Telecommunication Union (ITU) recommendations, if we con-
sider 22.5 GHz of frequency to establish SSLL, then -125 dBm
power is received for a 100 km link. Note that SSLL with
lower frequencies and distances have better energy-per-bit to
noise spectral density with fixing the gain of the transmitted
and received antennas. In Fig. 4, we show the energy-per-bit
to noise spectral density as a function of link distance. For
instance, the energy-per-bit to noise spectral density values
range between -2 and 19 dBm at 5 km for 60 GHz and 5.8
GHz, respectively. However, these values drop down to -48
and -28 dBm at 100 km.
2) Optical Link Budget: Another promising solution for
establishing SSLL is using FSO, as it can offer superior data-
rate compared to RF. Moreover, unlike RF communication,
7
FSO systems are easily expandable, light in weight, com-
pact, and easily deployable. Even in terms of bandwidth, the
permissible bandwidth can reach up to 20%of the carrier
frequency in RF systems; however, the utilized bandwidth at
an optical frequency is much higher even when it is taken to
be 1%of the carrier frequency [20]. Nevertheless, high-speed
optical links require a high directive beam that suffers from
ATP challenges, as mentioned earlier, and hence, restricted
to enable short-range SSLL. One possible solution to counter
the ATP issue is using photon-counting detector arrays at the
receiver that improves the signal acquisition for long-range
FSO communication [56]–[58].
FSO communication supports various binary and high-
level modulation schemes with different levels of power and
bandwidth efficiency for SSLL [20]. The most widely adopted
modulation format for optical SSLL is non-return-to-zero On-
Off Keying (OOK-NRZ) due to its easy implementation,
robustness, bandwidth efficiency, and direction detection fa-
cilitation. However, it imposes the constraint of an adaptive
threshold for getting the best results [59]. On the other
hand, M-Pulse Position Modulation (M-PPM) scheme does
not require an adaptive threshold, offering better average-
power efficiency, which in turn makes it a suotable choice
for deep-space communications [60]. However, in case of
limited bandwidth systems, increasing Mwould cause the
bandwidth efficiency to be substandard, and hence, high-level
schemes are more favorable. Besides M-PPM, optical sub-
carrier intensity modulation (SIM) does not require an adaptive
threshold as well. Furthermore, it provides more bandwidth
efficiency, less complicated design, and better bit error rate
(BER) than the M-PPM scheme. On the contrary, the SIM
scheme’s major disadvantage is the inferior power efficiency
as compared to OOK-NRZ and M-PPM [61]. According to
[62], homodyne BPSK is a recommended coherent modulation
scheme for SSLL because of its better communication and
tracking sensitivity. Moreover, it also gives complete protec-
tion from solar background noise. Another good candidate is
the differential phase-shift keying (DPSK) modulation scheme.
It considerably reduces power requirements and enhances
spectral efficiency than OOK-NRZ. However, it is complex
to design and hence expensive to implement [63].
Fig. 5: FSO-based LoS satellite-to-satellite link.
To calculate the optical link budget, we next consider light-
emitting diodes (LEDs) as transmitters and photodetectors
as receivers. The LEDs are assumed to use the OOK-NRZ
modulation scheme for enabling an optical SSLL. At the
receiver, the detector’s choice depends on various factors,
including cost, power level, the wavelength range of the
incident light, and the detector amplifier bandwidth. We refer
the interested readers to [64]–[66] for a detailed overview of
the types of photodetectors.
The generic LoS optical SSLL is illustrated in Fig. 5 where
dis the distance between satellites, αis the angle of incidence
with respect to the receiver axis, and βis the viewing angle
(irradiance angle) that describes the focus of the LED emitting
beam. In LoS optical links, the channel DC gain H(0) is
calculated as
H(0) = ((m+1)
2πd2Aocosm(β)Tfg(α) cos (α),: 0 ≤α≤αc
0,:α > αc
,
(6)
where mrepresents the order of Lambertian emission (i.e., a
quantity that expresses the radiation characteristics shape), Tf
is the filter transmission coefficient, g(α)is the concentrator
gain, and Aois the detector active area. The value of m
is related to the receiver field of view (FoV) concentrator
semi-angle αcat half illuminance of an LED Φ1/2as m=
−ln 2
ln(cos Φ1/2). Following the analysis in [67] and [68], an extra
concentrator gain is achieved by utilizing a hemispherical lens
with internal refractive index nas
g(α) = (n2
sin αc: 0 ≤α≤αc
0,:α > αc
.(7)
Hence, the received optical power (Pro) can be expressed as
Pro=H(0)Pt,(8)
At the receiver side, the electrical signal component can be
expressed by
S= (ξPro)2(9)
where ξis the photodetector responsivity. Therefore, the
required SNR at the receiver side can be determined given
that the total noise variance Nis the sum of noise variances
(shot noise σ2
sand thermal noise σ2
t), as
γreq =Eb
No
=[ξH (0)Pt]2
N
B
Rb
.(10)
Further evaluation of σ2
sand σ2
tcan be found in [67]. Based
on (10), Pbfor OOK scheme can be calculated as
Pb=1
2erfc 1
2√2√γreq .(11)
We now present a numerical link budget illustration by
considering a setup similar to the RF setup described earlier,
where the satellites orbit in LEO but with optical SSLL. The
parameters used for the simulations are mainly taken from
[67] and are listed in Table IV. In Fig. 6, we plot the received
power as a function of the concentrator FoV semi-angle. As
expected, Fig. 6 illustrates that as the distance between the
satellites increases, the received power decreases. Also, in the
case of a smaller concentrator angle, slightly more power is
received. Furthermore, in comparison with the RF case, the
8
TABLE IV: Parameters for the optical link budget
calculation.
Parameter Value
Transmitted power Pt2 W
Semi-angle at half power Φ1/230◦
Incidence angle α30◦
Irradiance angle β15◦
Detector responsivity ξ0.51
Refractive index of lens n1.5
Data rate Rb1 Mbps
Bandwidth B0.5 MHz
Detector active area Ao7.84 cm2
Absolute temperature T300 K
Filter transmission coefficient Tf1.0
LED wavelength λ656.2808 nm
received power using the optical technology is higher. For
example, at 5 km, the optical received power is approximately
-50 dBm; however, it swings between -70 and -90 dBm in the
RF scenario. Moreover, Fig. 7 presents the influence of the
concentrator FoV semi-angles on the energy-per-bit to noise
spectral density for different distances, where the performance
degrades with increasing the FoV of detectors and the distance
between satellites.
0 10 20 30 40 50 60 70 80 90 100
-95
-90
-85
-80
-75
-70
-65
-60
Fig. 6: The received power for different optical SSLL
distances.
B. Satellite-to-Satellite Links at Different Layers (SSLD)
Despite the fact that a single layer satellite network designed
by GEO, MEO, or LEO with P2P SSLL can offer multi-
media services to some degrees, many restrictions can affect
the performance of such a single layer satellite network. For
instance, a high accumulated delay is present in large constel-
lations due to multi-hops, and low stability is expected because
of the single-layer satellite network with planar topologies.
Moreover, repeated handovers lead to an increase in the prob-
ability of network routing and re-routing, creating congestions
0 10 20 30 40 50 60 70 80 90 100
-80
-70
-60
-50
-40
-30
-20
-10
Fig. 7: The energy-per-bit to noise spectral density for
different optical SSLL distances.
[69]. All the restrictions above harden the establishment and
maintenance of a single-layer satellite network.
Therefore, many studies on satellite-to-satellite links at
different layers (SSLD) exist in the literature. For instance, in
1997, [70] proposed the earliest two-layer satellite constella-
tion comprising of MEO and LEO satellites. The architecture
in [70] consists of both SSLL (among MEO satellites) and
SSLD (between LEO and MEO satellites). Consequently, [71]
proposed a similar two-layer MEO and LEO satellite network,
which included SSLL in each layer besides the SSLD. Their
network was designed to transmit short distance-dependent
services through SSLL, and relay long-distance traffics via
MEO satellites using SSLD. [14] introduces instead a more
complex multilayer satellite network architecture consisting of
GEO, MEO, and LEO satellites to improve capacity, reliability,
and coverage of satellite communication networks.
To implement such a multilayer satellite network, Japan
Aerospace Exploration Agency (JAXA) made various attempts
to develop a space data relay network for the next generation
of wireless communication systems. Moreover, various other
projects also tried to implement such multilayered satellite
networks with SSDL. Most of the recent works prefer to
use FSO for enabling satellite-to-satellite links at different
layers. One such project is Optical Inter-orbit Communications
Engineering Test Satellite (OICETS) “Kirari” by JAXA that
uses optical P2P links between satellites at different orbits.
Another similar project is “ARTEMIS” by ESA that also
uses optical links between the satellites at different altitudes
[72]–[75]. Some other similar projects are Alphasat TDP1
Sentinel-1A that uses FSO to relay data from GEO to LEO
[76] [77]. Moreover, recently, reference [78] propose a 20
Gbit/s-40 GHz OFDM based LEO-GEO optical system using
4-QAM modulation. Similarly, [79] presents a novel two-
layer satellite LEO/MEO network with optical links. On the
basis of the link quality, [80] introduces a novel QoS routing
protocol for LEO and MEO satellite networks. Furthermore,
Yan et al. discuss the topology analysis of two-layer links
9
in LEO/MEO satellite networks [81]. FSO communication
provides a promising solution to enable satellite-to-satellite
links at different altitudes because the radiated light beam is
not affected by the turbulence. However, FSO requires efficient
ATP mechanisms to provide reliable and stable links.
III. P2P LINKS IN HAP NETW OR KS
Unlike the satellites, HAPs operate at a much lower altitude,
i.e., around 20 km in the stratosphere above the earth’s surface.
The HAPs can provide ubiquitous connectivity in the operation
area since they can stay quasi-static in the air [96]–[99].
Numerous research projects use HAPs to enable connectivity,
especially in rural areas or in disaster-affected regions. One
such example is the Google Loon project, which aims to
provide Internet access in underserved areas. Table V presents
numerous HAPs projects that aim to develop aerial base
stations. Recently, HAPs-based wireless connectivity solutions
are promising due to the advances in the development of
lightweight materials and efficient solar panels that increase
the lifetime of HAPs and reduces the cost. Accordingly, a
set of inter-connected HAPs can be a transpiring solution to
provide Internet access and remote sensing in a broad region.
Therefore, it is interesting to discuss potential connectivity
solutions among HAPs that can lead to extended coverage
and perform backhauling.
A. HAP-to-HAP Links (HHL)
Early studies on establishing HAP-to-HAP Links (HHL)
and HAP backhauling mainly focus on radio communica-
tions. However, implementing RF links either for inter-HAP
communication or backhauling is not suitable for multiple
reasons, e.g., such links require high bandwidth and high
transmit power for long-range communication [48]. Besides,
wireless communication links at a higher RF frequency band
are severely affected by environmental impediments, such
as rain attenuation. Irrespective of these challenges, various
works studied RF-based HHL and backhaul links [100]–[106].
For instance, [100] proposes a backhaul link between the
HAP with WiMAX payload and the customer premises on
the ground. Consequently, [102] investigates digital video
broadcasting protocol (DVB-S2) for the backhauling to the
ground station by using HAPs, which shows that the BER
is low compared to WiMAX at lower SNR. [103] highlights
the effects of weather conditions on the performance of
HAPs backhaul links. Moreover, recently, [104] optimizes the
cell configuration for a high-speed HAPs system by using a
genetic algorithm that also tries to minimize the total power
consumption.
Besides HAPs backhauling, interconnecting the HAPs re-
quire high-speed communication links. Therefore, unlike the
HAP-to-ground links, which mainly uses RF communication,
establishing inter-HAP links prefer to use FSO communication
[107], [108]. The FSO links are vulnerable to weather condi-
tions, such as clouds and fog. However, the HAPs are operat-
ing above the clouds; thus, FSO links are less affected at such
an altitude. For example, [109] proposes a 500 km inter-HAP
FSO link at 20 km of altitude, achieving 384 Mbps of data
rate with 10−6BER. Likewise, [110] performs BER analysis
for FSO-based inter-HAP links in the presence of atmospheric
turbulence, where the BER increases with an increase in the
scintillation index and link distance. In order to evaluate the
performance of FSO-based HHL, it is important to develop
accurate channel models that account for various losses such
as geometrical loss, turbulence, inhomogeneous temperature,
and pointing error. Geometrical loss mainly occurs due to
the spreading of light resulting in less power collected at the
receiver. On basis of the path length d, radius of the receiver
aperture r, and divergence angle α, the geometrical loss can
be represented as
Lg=4πr2
π(αd)2.(12)
Similarly, the estimation of turbulence loss requires to measure
the turbulence strength with changing refractive index param-
eter n2(h)at various altitudes. Various empirical models, such
as Hufnagel-Valley (H-V) model are used to estimate n2(h).
On the basis of (H-V) model, n2(h)as a function of altitude
(h) is measured as
n2(h)=0.00594 ν
272(10−5h)10 exp −h
1000(13)
+2.7×10−16 exp −h
1500+Kexp −h
100,
where νis the wind speed and K= 1.7×10−14m−2/3
is constant. Based on (13), the turbulance loss in dB’s is
calculated as
Lt= 2s23.17 2π
λ1097/6
n2(h)d11/6.(14)
Additionally, the pointing loss occurs due to numeorus reasons
such as wind, jitter, turbulence, and vibration of HAPs. The
pointing error can result in a link failure or reduces the
amount of power received at the receiver resulting in a high
BER. Therefore, it is crucial to model the pointing error
both in azimuth and elevation. There are various statistical
distributions in the literature to model the pointing error for
FSO communication, such as Rayleigh distribution [111], Hoyt
distribution [112], Rician distribution [113], and Beckmann
distribution [114]. In case when the pointing error is modeled
as Gaussian distribution, the radial error angle e=pθ2+φ2
is the function of elevation (θ) and azimuth (φ) angles.
Considering that θand φare zero-mean i.i.d processes with
variance σ, then the pointing error follows Rician distribution
as follows
f(θ, β) = θ
σ2exp −θ2+β2
2σ2I0θβ
σ2,(15)
where βis the angle bias error from the center and I0(·)is
the zeroth-order Bessel function. In case when β= 0, (15)
leads to Rayleigh distribution function, given as
f(θ) = θ
σ2exp −θ2
2σ2.(16)
The pointing error for FSO-based inter-HAP links can be
mitigated by increasing the receiver FoV, using multiple beam
10
TABLE V: List of various projects on HAPs.
Project Type Technology Link Type Organization Description
SHARP [82] Aerodynamic Microwave HAP-Ground Communications
Research Centre
(CRC)
It goes to prove successful one-hour commu-
nication flight time.
Pathfinder,
Centurion, and
Helios [83]
Aerodynamic RF HAP-Ground NASA This project consists of a solar powered aero-
dynamic HAP providing high-definition TV
(HDTV) transmissions and 3G communication
services.
SkyNet [84]–
[86]
Aerostatic-
(Airship)
RF HAP-Ground JAXA SkyNet promotes future high-speed wireless
communications by using a 200 m length air-
ship that can operate for up to 3 years.
CAPANINA
[87]
Aerostatic-
(Balloon)
Optical and
RF
HAP-Ground University of
York
This project provides enhance broadband ac-
cess for both urban and rural communities
in Europe, demonstrating data transmission of
1.25 Gbps.
X-station [88] Aerostatic-
(Airship)
RF HAP-Ground StratXX X-station airship can stay in the air for around
an year providing various communication ser-
vices, such as TV and radio broadcasting,
mobile telephony, VoIP, remote sensing, and
local GPS.
Elevate [89] Aerostatic-
(Balloon)
RF HAP-Ground Zero 2 Infinity Elevate balloons can lift payloads up to 100 kg
to test and validate novel technologies in the
stratosphere.
Loon [90] Aerostatic-
(Balloon)
Optical HAP-Ground
and IHAP
Alphabet Inc. The aim of this project is to connect people
globally using a network of HAPs with each
balloon having 40 km of coverage radius. The
balloons in this project can stay in the air for
223 days.
Zephyr S [91] Aerodynamic RF HAP-Ground Airbus Project Zephyr S can lift a payload of up to
12 kg and can flight continuously for around
100 days, aiming to connect the people in
underserved areas, achieving 100 Mbps.
Aquila [92] Aerodynamic RF HAP-Ground Facebook Similar to Zephyr S, the goal of Aquila was to
provide broadband coverage in remote areas.
Stratobus [93] Aerostatic-
(Airship)
Optical HAP-Ground
and IHAP
Thales Alenia
Space
Unlike other HAPs, Stratobus can support
heavy payload, i.e., up to 450 kg and stay
almost static in the stratosphere for a longer
time (up to 5 years), providing 4G/5G com-
munication services.
HAWK30 [94] Aerodynamic mmWave HAP-Ground SoftBank Corp. This project consists of HAPs with each having
100 km of coverage, aiming to ground users,
UAVs, IoT devices.
PHASA-35
[95]
Aerodynamic RF HAP-Ground Prismatic Project PHASA-35 can support up to 35 kg of
payload and can fly continuously for an year
to provide 5G communication services.
transmissions, hybrid RF/FSO, and adaptive optics [115].
In the literature, various statistical channel models can be
found that models the propagation characteristics of FSO
communication. For example, [116] propose a gamma-gamma
distribution for a laser link in the presence of turbulence.
[117] uses log-normal distribution to model the FSO links
with fluctuations. These statistical fading models can estimate
the scintillation index for FSO links and help in analyzing
these links. For example, the log-normal distribution estimates
well the weak turbulence; however, it underestimates the
distribution’s tails and peaks. In contrast, exponential channel
distribution fits well for a strong turbulence region but is
not consistent for weak turbulence. Nevertheless, the gamma-
gamma channel model works well for both weak and strong
turbulence regimes [116]. Similarly, Malaga distribution also
fits well for a wider range of turbulence effects where log-
normal and gamma-gamma distributions are its special cases.
In the case of a gamma-gamma channel model, the probability
distribution function (PDF) for the irradiance Ircan be written
as
fIr(I) = 2(¯α¯
β)¯α+¯
β
2
Γ(¯α)Γ( ¯
β)Ibaralpha+¯
β
2J¯α−¯
β2q¯α¯
βI (17)
where ¯αand ¯
βare the fading parameters for turbulence,
Γ(·)is the gamma function, and J(·)is the second order
modified Bessel function. Based on the values of ¯αand ¯
β, the
scintillation index for gamma-gamma model can be written as
σI=1
¯α+1
¯
β+1
¯α¯
β(18)
Note that the effect of turbulence can be mitigated by using
aperture averaging, i.e., increasing the aperture size reduces
11
the fluctuations leading to a lower scintillation index [114].
The interested readers are referred to [118] for various FSO
channel models that can be used for establishing inter-HAP
links.
In the presence of the impediments mentioned above, re-
searchers have studied the performance HAPs regarding cov-
erage and capacity. Nevertheless, most of the existing works
study HAP-to-ground links using geometrical and statistical
models [119]. For instance, [120] investigates BER perfor-
mance for hybrid WiMAX and HAP-based connectivity solu-
tions for ground users. [121] performs the capacity analysis
for a MIMO-based communication link between the HAP and
a high-speed train, which shows that although there is a strong
LoS component, the channel is still ill-conditioned. Similarly,
[122] designs HAPs-based backhaul link using FSO in the
presence of turbulence, achieving 1.25 Gbps with BER of less
than 10−9. Consequently, [123] studies a 3D channel model to
see the impact of distance among antennas in a MIMO-HAP
system, where the channel is affected by the distribution of
scatters, array configuration, and Doppler spread. Moreover,
[124] investigates interference for ground users with two
HAPs, showing that better performance is achieved if the users
are spatially well separated. In [125], the authors improve the
capacity of HAP systems by using mmWave frequencies. [125]
also evaluates ground users’ capacity regarding the angular
separation between the ground users and HAPs. Furthermore,
[125] analyze the coverage of HAPs operating at 48 GHz
and 28 GHz frequencies discussing various crucial system
parameters, including beam type and frequency reuse for
cell planning. [126] focuses on the deployment of HAPs to
characterize the HAP-to-ground link in terms of path loss and
maximizes the on-ground coverage.
Moreover, [127] investigates the use of relays in the pres-
ence of turbulence and pointing errors for multi-hop FSO that
can be used for establishing inter-HAP links. [127] analyzes
amplify-and-forward relaying with channel state information
and fixed-gain relays regarding signal-to-interference-plus-
noise ratio (SINR) and coverage probability. Consequently,
[128] derives the closed-form expression of BER and channel
capacity showing the effects of pointing errors and beam
wandering for FSO-based inter-HAP links. Michailidis et al.
further investigates hybrid triple-hop RF-FSO-RF links for
HAPs based communication systems where the two HAPs are
connected through FSO while the HAP-to-ground link is RF
[129]. Fig. 8 illustrates such a hybrid RF-FSO architecture
where FSO can be used in good weather conditions to achieve
higher data rates while RF can be utilized in bad weather
conditions and in the absence of LoS.
B. Handover between HAPs
The HAPs in the stratospheric atmosphere can be affected
by the airflow, resulting in a different footprint on the ground.
Therefore, it is crucial to design handover schemes for the
ground users to maintain the communication link. Handover
in HAP networks is the process of transferring the commu-
nication link between cells to avoid the channel’s instability.
This process usually occurs when there are massive differences
between cell sizes in HAP extended coverage scenarios [48].
Many works in the literature discuss handover schemes for
a stand-alone HAP or between HAP networks [79], [130]–
[134]. In [79], [132], [133], the authors focus on minimizing
the traffic difference between cells during the data transfer,
considering the HAP travel direction, the adaptive modulation,
and cells cooperation, respectively. On the other hand, Lim et
al. suggest an adaptive soft handover algorithm using both
the platform’s downlink output power and individual base
stations in [130]. In [131], the authors discuss the influence
of platform movement on handover. Moreover, a handover
decision algorithm based on prediction, using the time series
analysis model with an adaptive threshold, is designed in
[134]. We wish to finally mention that most the link budget
illustrations of P2P links in satellite networks discussed in
the previous section also apply to HAP networks, and so
we choose not to explicitly describe them in the text for
conciseness.
IV. INT EG RATED SATELLITE-HAP COMMUNICATION
NET WORKS
6G wireless communication systems envision to provide
broadband services in underserved areas with reasonable costs.
Satellite networks are one possible enabler of such a vision due
to their large footprints and their capabilities to provide ubiq-
uitous coverage to remote areas. Recently, mega-constellations
of small satellites in LEO gain interest in academia and indus-
try to enable broadband services worldwide [135]. Moreover,
the development of integrated satellite-HAPs-LAPs networks
can further improve the coverage, reliability, and scalability of
6G wireless communication systems [136]–[138]. A potential
integrated spatial network consists of spatial nodes at the
same or different altitudes connected via either RF or optical
links. For example, satellite networks can provide RF/optical
backhauling for HAPs and LAPs.
Recently, various research works are devoted to the vision
of integrated spatial networks. For example, [139] proposes an
integrated spatial and terrestrial system consisting of satellites
with mounted BSs, UAVs, and ground vehicles. Their solution
is based on densification to increase the network capacity in
the demand area. However, the proposed architecture in [139]
is a function of several challenges, such as interoperability,
resource allocation, and network management for a highly dy-
namic environment. To this end, [140] develops SAGECELL,
a software-defined integrated spatial/terrestrial moving cell
solution. The SDN-based approach results in flexible resource
allocation with centralized network management. Moreover,
[141] proposes an integrated satellite-terrestrial access network
where the LEO-based small cells coordinate with small terres-
trial cells to improve wireless access reliability and flexibility.
However, this approach requires ultra-dense deployment of
LEO satellites and also ignores HAPs and LAPs. Zhu et al.
propose a cloud-based integrated satellite-terrestrial network
where both the satellite and ground BSs are connected to
a common baseband processing system that performs in-
terference mitigation, cooperative transmission, and resource
management [142].
12
Fig. 8: An architecture of HAPs network with P2P HAP-to-HAP and backhauling links.
Unlike the works mentioned above, [143] introduces a
heterogeneous spatial network consisting of satellites, HAPs,
and LAPs. The backbone network entities are connected
via laser links and the access network, allowing the user
to enter the spatial network using microwave links. Several
industrial projects have been launched to realize such an
architecture. For example, Integrated Space Infrastructure for
global Communication (ISICOM) [144] and Transformational
Satellite Communications System (TSAT) [145] aim to pro-
vide global communication, covering oceans, ground, and
space. Moreover, various works investigate the communication
link between HAPs and satellites. For instance, [146] explores
optical HAP-to-LEO links where the reliability of the link
degrades at low elevation angles. Similarly, [147] proposes
a HAP-based relaying for FSO communication between the
ground and LEO satellites. Thanks to the HAP-based relaying,
it increases the power gain by 28 dB at BER of 10−9[147].
V. FUTURE RESEARCH DIRECTIONS
On the basis of the literature we reviewed, this section
outlines numerous promising future research challenges for in-
tegrated spatial networks. Since the studies on these complex,
large-scale spatial networks are still at initial stages, various
problems need further investigation. In the following, we point
out to some of these open research issues.
A. Network Optimization
Network optimization for an integrated spatial network is
much more complicated than a stand-alone terrestrial or an
aerial network because of the diverse characteristic of spatial
nodes at each layer. Therefore, novel optimization techniques
are required to consider various network characteristics, such
as cost, mobility, energy efficiency, spectrum efficiency, and
user experience. Recently, the use of artificial intelligence
is gaining interest in optimizing such large-scale networks.
For instance, [148] employs a deep neural network model to
optimize wireless networks’ energy consumption. Similarly,
[149] uses reinforcement learning with a Bayesian network to
maximize the throughput of a D2D network. Likewise, [150]
targets to improve mobility robustness using Q-learning for
cellular networks. Recently, [151] uses artificial intelligence
to optimize integrated spatial and ground networks regarding
traffic control, resource allocation, and security. However, the
existing works on optimization for spatial networks remain
relatively limited, and so advanced joint optimization tech-
niques need to be developed to address various issues of spatial
networks, such as cost, spectrum utilization, security, traffic
offloading, and energy efficiency.
B. Intelligent Offloading
There has been a plethora of work on traffic offloading
in different wireless networks, including satellite, UAVs, and
terrestrial networks [141]. With the recent advancements in
13
integrated spatial networks, new possibilities for traffic of-
floading arise. Nevertheless, resource management and coordi-
nated traffic offloading in such an integrated network are more
complicated than a standalone non-terrestrial or terrestrial
network [152]. For example, satellite connections have large
latency, which means low QoE compared to terrestrial links.
Concurrently, satellite links are more appealing for continued
services and seamless connectivity due to its wider footprint.
Recently, [153] proposes a latency-aware scheme for traffic
offloading in integrated satellite-terrestrial networks where the
URLLC requirement is satisfied for traffic offloading to the
terrestrial backhaul. In contrast, eMBB data is offloaded to
the satellites as eMBB traffic does not have always a stringent
delay requirement. Moreover, intelligent traffic offloading in
integrated spatial-terrestrial networks can be enabled using
SDN technology that can separate the data and network
plans [154]. Also, based on link characteristics, such as
cost, reliability, and capacity, multiple options can offload the
data. Therefore, it is interesting to investigate different traffic
offloading schemes for integrated spatial-terrestrial networks
to make optimum offloading decisions.
C. Smart Platforms
Intelligent reflecting surfaces, also known as smart surfaces
(SS) have emerged as promising 6G wireless communication
technology. These smart surfaces consist of flexible metamate-
rials that allow them to passively/actively reflect the received
signals improving the communication channel’s quality [155].
Considering numerous smart surfaces’ opportunities, it is well-
suited for the spatial platforms, including satellites, HAPs,
and UAVs [156]. For instance, [157] proposes SS-assisted
THz communication links for LEO satellite networks where
SS improve the SNR of the received signal. Similarly, [158]
investigates the link budget analysis for communication in SS-
assisted aerial platforms. SS-assisted spatial platforms offer
several advantages, including energy efficiency, improved cov-
erage, and lower system complexity. Despite these benefits,
the research on SS-assisted spatial platforms is in infancy and
needs further investigation.
D. Energy Efficiency
The limited power supply of spatial platforms requires to
use the on-board energy efficiently. Unlike terrestrial networks
where most of the energy is consumed in communication,
spatial networks are also affected by radiations, space/aerial
environment, and different propagation channels [159]. One
way to reduce spatial platforms’ power consumption is to
design power amplifiers with a low peak-to-average power
ratio (PAPR). Novel techniques such as non-orthogonal wave-
forms can be investigated to reduce the PAPR. Moreover,
spatial platforms’ energy consumption can also be reduced by
using new networking technologies, such as SDN and NFV.
In [160], the authors reveal that significant energy gain can
be accomplished for integrated spatial-terrestrial networks by
splitting the control and data plans using SDN. Furthermore,
energy harvesting techniques need to be explored to make
spatial networks green and environment friendly.
E. Novel Multiple Access Schemes
Several multiple access schemes, such as space-division
multiple access (SDMA) and non-orthogonal multiple access
(NOMA), are promising for multiplexing in aerial networks.
However, the gain of SDMA and NOMA is limited because
they depend on environmental conditions. Therefore, [161]
introduces rate-splitting multiple access (RSMA), which has
better spectral efficiency for an integrated network. In the
context of integrated spatial-terrestrial networks, RSMA can
be employed horizontally at one of the layers or vertically
at each layer [162]. The management of RSMA can be
performed centrally (if a central controller manages a layer)
or in a distributed fashion (if layers are separately managed).
Nevertheless, the investigation of RSMA in such scenarios is
missing in the literature and needs the researchers’ attention.
F. Distributed Spatial Networks
The spatio-temporal variations of the flying platforms and
their relative positioning are critical aspects of the ground-
level communications metrics. While satellites move in pre-
determined constellations which typically consist of comple-
mentary orbital planes [4], HAPs are relatively stationary
within the stratospheric layer [48]. LAPs (e.g., UAVs), on the
other hand, are distributed platforms capable of dynamically
adjusting their locations based on both the underlying ground-
demand, and the heterogeneous nature of the wireless network;
see [163] and references therein. Automating LAPs positioning
becomes, therefore, an important aspect of terrestrial-aerial
networks design so as to improve the overall system quality-
of-service. From an end-to-end system-level perspective, the
provisioning of the spatio-temporal variations of the network
(e.g., data traffic, user-locations, etc.) and the positioning of
the aerial networks (e.g., UAVs dynamic positioning, satellite
constellations design, HAPs placement, etc.) becomes crucial
both to capture the instantaneous and the long-term network
metrics, and to optimize the network parameters accordingly.
A future research direction is, therefore, to enable the real-time
operations of such distributed systems, mainly LAPs-to-LAPs
and LAPs-to-ground, through the accurate modeling of the
networks variations, and through invoking the proper online
distributed optimization for real-time data processing.
VI. CONCLUSIONS
Spatial networks are emerging as major enablers for next-
generation wireless communications systems. Through their
invigorating capabilities in providing connectivity solutions,
improving remote areas’ coverage, and increasing the data
capacity in metropolitan regions, such spatial networks are
expected to offer global Internet for all and, at the same
time, provide terrestrial wireless backhaul solutions. Assessing
the true benefits arising from integrating various single-layer
networks at different altitudes (such as satellites, HAPs, and
LAPs) remains, however, subject to several physical hur-
dles. Unlike terrestrial networks, high latency, constrained
resources, mobility, and intermittent links are major spatial
network issues, and so it becomes vital to study the intercon-
necting P2P links among various layers of spatial networks.
14
To this end, this paper surveys the state-of-the-art on enabling
P2P links in different layers of spatial networks. The paper first
introduces spatial networks’ background, including satellite,
HAPs, and LAPs networks, and presents various exciting
projects on the topic. Then, we explain two different solutions,
i.e., RF and FSO, for connecting the satellites in a single
orbit or at different orbits. We also present the link budget
analysis for both RF and FSO-based satellite-to-satellite links.
Furthermore, we present the studies regarding RF and FSO
for enabling HAP-to-HAP links and further explore the re-
search on performance analysis of HAP networks. Afterward,
we present the literature on integrated terrestrial and non-
terrestrial networks as a means to enable next-generation wire-
less communication systems. Finally, we identify numerous
future research directions, including network optimization,
intelligent offloading, smart platforms, energy efficiency, mul-
tiple access schemes, and distributed spatial networks. Up
to the authors’ knowledge, this is the first paper of its kind
that surveys P2P links for a multi-layered spatial network in
light of 6G large-scale complex networks. Many of the paper
insights intend at enabling the establishment of P2P links in
future integrated spatial networks.
REFERENCES
[1] G. Hernandez and V. Weber, “From people to things:
Building global connectivity - OECD observer,” https:
//oecdobserver.org/news/fullstoryphp/aid/5587/Frompeople to things:
Building global connectivity .html, 2020.
[2] S. Dang, O. Amin, B. Shihada, and M.-S. Alouini, “What should 6G
be?” Nature Electronics, vol. 3, no. 1, pp. 20–29, 2020.
[3] E. Yaacoub and M.-S. Alouini, “A key 6G challenge and opportu-
nity—connecting the base of the pyramid: A survey on rural connec-
tivity,” Proc. of the IEEE, vol. 108, no. 4, pp. 533–582, 2020.
[4] N. Saeed, A. Elzanaty, H. Almorad, H. Dahrouj, T. Y. Al-Naffouri, and
M.-S. Alouini, “Cubesat communications: Recent advances and future
challenges,” IEEE Commun. Surveys Tuts., vol. 22, pp. 1839 – 1862,
Apr. 2020.
[5] H. Saarnisaari, S. Dixit, M.-S. Alouini, A. Chaoub, M. Giordani,
A. Kliks, M. Matinmikko-Blue, N. Zhang, A. Agrawal, M. Andersson
et al., “A 6G white paper on connectivity for remote areas,” arXiv
preprint arXiv:2004.14699, 2020.
[6] X. Huang, J. A. Zhang, R. P. Liu, Y. J. Guo, and L. Hanzo, “Airplane-
aided integrated networking for 6G wireless: Will it work?” IEEE Veh.
Techno. Mag., vol. 14, no. 3, pp. 84–91, 2019.
[7] A. A. Abu-Arabia, R. Hakimi et al., “Performance of 5G services
deployed via HAPS system,” in IEEE 13th Int. Conf. on Telecommun.
Sys., Serv., and Apps., (TSSA). IEEE, 2019, pp. 168–172.
[8] J. Qiu, D. Grace, G. Ding, M. D. Zakaria, and Q. Wu, “Air-ground
heterogeneous networks for 5G and beyond via integrating high and
low altitude platforms,” IEEE Wireless Commun., vol. 26, no. 6, pp.
140–148, 2019.
[9] F. Rinaldi, H. L. Maattanen, J. Torsner et al., “Non-terrestrial networks
in 5G beyond: A survey,” IEEE Access, vol. 8, pp. 165 178–165 200,
2020.
[10] R. Czichy, “Inter-satellite links for advanced space networks,” in 18th
Int. Commun. Satellite Sys. Conf. and Exhibit, 2000, p. 1261.
[11] N. Saeed, T. Y. Al-Naffouri, and M.-S. Alouini, “Wireless
communication for flying cars,” KAUST Repository, 2020. [Online].
Available: https://repository.kaust.edu.sa/handle/10754/662814
[12] N. Saeed, T. Y. Al-Naffouri, and M.-S. Alouini, “Around the world of
IoT/Climate monitoring using internet of X-things,” IEEE Internet of
Things Mag., vol. 3, no. 2, pp. 82–83, 2020.
[13] M. Barthelemy, “Discussion: Social and spatial networks,” Les nou-
velles de l’arch´
eologie, no. 135, pp. 51–61, 2014.
[14] I. F. Akyildiz, E. Ekici, and M. D. Bender, “Mlsr: a novel routing
algorithm for multilayered satellite ip networks,” IIEEE/ACM Trans.
Netw., vol. 10, no. 3, pp. 411–424, 2002.
[15] X. Cao, P. Yang, M. Alzenad, X. Xi, D. Wu, and H. Yanikomeroglu,
“Airborne communication networks: A survey,” IEEE J. Sel. Areas
Commun., vol. 36, no. 9, pp. 1907–1926, 2018.
[16] C. P. Dettmann, O. Georgiou, and P. Pratt, “Spatial networks with
wireless applications,” Comptes Rendus Physique, vol. 19, no. 4, pp.
187 – 204, 2018.
[17] Ares, “Inter-satellite links and ARES capabilities,” http:
//www.ares-consortium.org/It/wp-content/uploads/2016/10/13.10.16-
ISL-ARES- Capabilities-.pdf, online; accessed 14 Sep. 2020.
[18] J. Yang, J. Boyd, D. Laney, and J. Schlenzig, “Next generation
half-duplex common data link,” in IEEE Military Commun. Conf.,
(MILCOM), 2007, pp. 1–7.
[19] J. C. Juarez, D. W. Young, R. A. Venkat, D. M. Brown, A. M. Brown,
R. L. Oberc, J. E. Sluz, H. A. Pike, and L. B. Stotts, “Analysis of
link performance for the FOENEX laser communications system,” in
Atmospheric Propagation IX, vol. 8380, 2012, p. 838007.
[20] H. Kaushal and G. Kaddoum, “Optical communication in space:
Challenges and mitigation techniques,” IEEE Commun. Surveys Tuts.,
vol. 19, no. 1, pp. 57–96, 2016.
[21] W. D. Williams, M. Collins, D. M. Boroson et al., “RF and optical
communications: A comparison of high data rate returns from deep
space in the 2020 timeframe,” NASA, Report, 2007.
[22] H. Henniger and O. Wilfert, “An introduction to free-space optical
communications,” Radioengineering, vol. 19, no. 2, 2010.
[23] A. Douik, H. Dahrouj, T. Y. Al-Naffouri, and M.-S. Alouini, “Hybrid
radio/free-space optical design for next generation backhaul systems,”
IEEE Trans. Commun., vol. 64, no. 6, pp. 2563–2577, 2016.
[24] Z. Gibalina and V. Fadeev, “Optical inter-satellite link in comparison
with RF case in cubesat system,” J. of Radio Electronics, no. 10, 2017.
[25] P. R. Horkin and K. A. Olds, “Future optical ISL characteristics from
the perspective of large commercial constellations,” Space Commun.,
vol. 15, no. 2, pp. 83–87, 1998.
[26] P. Zablocky, “100 Gb/s RF backbone,” https://www.darpa.mil/program/
100-gb- s-rf- backbone, online; accessed 14 Sep. 2020.
[27] A. Malinowski and R. J. Zieli´
nski, “High altitude platform: Future of
infrastructure,” Int. J. of Electronics and Telecommun., vol. 56, no. 2,
pp. 191–196, 2010.
[28] DoD, “FSO and RF integrated aerial communications (FaRIA-C),”
https://www.sbir.gov/sbirsearch/detail/1531869, online; accessed 14
Sep. 2020.
[29] L. B. Stotts, L. C. Andrews, P. C. Cherry, J. J. Foshee, P. J. Kolodzy,
W. K. McIntire, M. Northcott, R. L. Phillips, H. A. Pike, B. Stadler
et al., “Hybrid optical RF airborne communications,” Proc. of the IEEE,
vol. 97, no. 6, pp. 1109–1127, 2009.
[30] K. Maine, C. Devieux, and P. Swan, “Overview of IRIDIUM satellite
network,” in Proc. of WESCON, 12 1995, pp. 483–490.
[31] ESA, “a world first: data transmission between european satellites using
laser light,” http://www.esa.int/Applications/Telecommunications
Integrated Applications/A world first Data transmission between
European satellites using laser light, online; accessed 14 Sep. 2020.
[32] B. Thompson, “Falcon fast, far, and first,” online; accessed 14 Sep.
2020. [Online]. Available: https://www.wpafb.af.mil/News/Article-
Display/Article/400176/falcon-fast-far-and-first/
[33] DoD, “Laser for airborne communications (LAC),” https:
//www.sbir.gov/node/561338, online; accessed 14 Sep. 2020.
[34] ——, “cognitive uhf radio for enhanced gps crosslinks,” https://
www.sbir.gov/sbirsearch/detail/870305, online; accessed 14 Sep. 2020.
[35] QB50, “Project qb50,” https://www.qb50.eu/, online; accessed 14 Sep.
2020.
[36] F. Sansone, A. Francesconi, R. Corvaja, G. Vallone, R. Antonello,
F. Branz, and P. Villoresi, “LaserCube optical communication terminal
for nano and micro satellites,” Acta Astronautica, vol. 173, pp. 310 –
319, 2020.
[37] S. Karapantazis and F. Pavlidou, “Broadband communications via high-
altitude platforms: A survey,” IEEE Commun. Surveys Tuts., vol. 7,
no. 1, pp. 2–31, 2005.
[38] F. Fidler, M. Knapek, J. Horwath, and W. R. Leeb, “Optical commu-
nications for high-altitude platforms,” IEEE J. Sel. Topics Quantum
Electron., vol. 16, no. 5, pp. 1058–1070, 2010.
[39] J. Mukherjee and B. Ramamurthy, “Communication technologies and
architectures for space network and interplanetary internet,” IEEE
Commun. Surveys Tuts., vol. 15, no. 2, pp. 881–897, 2013.
[40] ˙
Ilker Bekmezci, O. K. Sahingoz, and S¸ amil Temel, “Flying ad-hoc
networks (FANETs): A survey,” Ad Hoc Netw., vol. 11, no. 3, pp.
1254 – 1270, 2013.
15
[41] Y. Chen, H. Zhang, and M. Xu, “The coverage problem in UAV
network: A survey,” in Fifth Int. Conf. on Comput., Commun. and Netw.
Techno., (ICCCNT), 2014, pp. 1–5.
[42] L. Gupta, R. Jain, and G. Vaszkun, “Survey of important issues in UAV
communication networks,” IEEE Commun. Surveys Tuts., vol. 18, no. 2,
pp. 1123–1152, 2016.
[43] R. Radhakrishnan, W. W. Edmonson, F. Afghah, R. M. Rodriguez-
Osorio, F. Pinto, and S. C. Burleigh, “Survey of inter-satellite com-
munication for small satellite systems: Physical layer to network layer
view,” IEEE Commun. Surveys Tuts., vol. 18, no. 4, pp. 2442–2473,
2016.
[44] I. K. Son and S. Mao, “A survey of free space optical networks,” Digital
Commun. and Netw., vol. 3, no. 2, pp. 67–77, 2017.
[45] A. A. Khuwaja, Y. Chen, N. Zhao, M. Alouini, and P. Dobbins, “A
survey of channel modeling for UAV communications,” IEEE Commun.
Surveys Tuts., vol. 20, no. 4, pp. 2804–2821, 2018.
[46] Y. Peng, T. Dong, R. Gu, Q. Guo, J. Yin, Z. Liu, T. Zhang, and Y. Ji,
“A review of dynamic resource allocation in integrated satellite and
terrestrial networks,” in 2018 Int. Conf. on Networking and Netw. App.,
(NaNA), 2018, pp. 127–132.
[47] A. Fotouhi, H. Qiang, M. Ding, M. Hassan, L. G. Giordano, A. Garcia-
Rodriguez, and J. Yuan, “Survey on UAV cellular communications:
Practical aspects, standardization advancements, regulation, and secu-
rity challenges,” IEEE Commun. Surveys Tuts., vol. 21, no. 4, pp. 3417–
3442, 2019.
[48] S. C. Arum, D. Grace, and P. D. Mitchell, “A review of wireless
communication using high-altitude platforms for extended coverage
and capacity,” Computer Commun., 2020.
[49] O. Kodheli, E. Lagunas, N. Maturo, S. K. Sharma et al., “Satellite
communications in the new space era: A survey and future challenges,”
arXiv, 2020. [Online]. Available: https://arxiv.org/abs/2002.08811
[50] M. R. Bhatnagar, “Performance evaluation of decode-and-forward
satellite relaying,” IEEE Trans. Veh. Techno., vol. 64, no. 10, pp. 4827–
4833, 2015.
[51] Z. Yoon, W. Frese, and K. Briess, “Design and implementation of a
narrow-band intersatellite network with limited onboard resources for
IoT,” Sensors, vol. 19, no. 19, p. 4212, 2019.
[52] B. G. Evans, P. T. Thompson, G. E. Corazza, A. Vanelli-Coralli, and
E. A. Candreva, “1945-2010: 65 years of satellite history from early
visions to latest missions,” Proc. of the IEEE, vol. 99, no. 11, pp.
1840–1857, 2011.
[53] O. Popescu, “Power budgets for cubesat radios to support ground
communications and inter-satellite links,” IEEE Access, vol. 5, pp.
12 618–12 625, 2017.
[54] NASA, “Intersatellitelink (ISL) application to commercial com-
munications satellites volume II,” https://ntrs.nasa.gov/archive/nasa/
casi.ntrs.nasa.gov/19870010120, online; accessed 14 Sep. 2020.
[55] B. Sklar et al.,Digital communications: fundamentals and applications,
2001.
[56] M. S. Bashir and M.-S. Alouini, “Signal acquisition with photon-
counting detector arrays in free-space optical communications,” IEEE
Trans. on Wireless Commun., vol. 19, no. 4, pp. 2181–2195, 2020.
[57] ——, “Free-space optical MISO communications with an array of
detectors,” 2020. [Online]. Available: https://arxiv.org/abs/2009.00380
[58] ——, “Robust beam position estimation with photon counting
detector arrays in free-space optical communications,” 2020. [Online].
Available: https://repository.kaust.edu.sa/handle/10754/665446
[59] A. Jain, R. K. Bahl, and A. Banik, “Demonstration of RZ-OOK mod-
ulation scheme for high speed optical data transmission,” in Eleventh
Int. Conf. on Wireless and Optical Commun. Netw., (WOCN), 2014,
pp. 1–5.
[60] H. Hemmati, Deep space optical communications. John Wiley &
Sons, 2006, vol. 11.
[61] N. D. Chatzidiamantis, A. S. Lioumpas, G. K. Karagiannidis, and
S. Arnon, “Adaptive subcarrier PSK intensity modulation in free space
optical systems,” IEEE Trans. Commun., vol. 59, no. 5, pp. 1368–1377,
2011.
[62] R. Lange and B. Smutny, “Homodyne BPSK-based optical inter-
satellite communication links,” in Free-Space Laser Communication
Technologies XIX and Atmospheric Propagation of Electromagnetic
Waves, S. Mecherle and O. Korotkova, Eds., vol. 6457, 2007, pp. 19
– 27.
[63] W. A. Atia and R. S. Bondurant, “Demonstration of return-to-zero sig-
naling in both OOK and DPSK formats to improve receiver sensitivity
in an optically preamplified receiver,” in IEEE LEOS Annual Meeting
Conf. Proc., vol. 1, 1999, pp. 226–227.
[64] L. Wood, W. Ivancic, and K.-P. D ¨
orpelkus, “Using light-emitting diodes
for intersatellite links,” in IEEE Aerospace Conf., 2010, pp. 1–6.
[65] J. M. Kahn and J. R. Barry, “Wireless infrared communications,” Proc.
of the IEEE, vol. 85, no. 2, pp. 265–298, 1997.
[66] O. Kharraz and D. Forsyth, “Performance comparisons between PIN
and APD photodetectors for use in optical communication systems,”
Optik, vol. 124, no. 13, pp. 1493–1498, 2013.
[67] D. N. Amanor, W. W. Edmonson, and F. Afghah, “Intersatellite
communication system based on visible light,” IEEE Trans. Aerosp.
Electron. Syst., vol. 54, no. 6, pp. 2888–2899, 2018.
[68] I. E. Lee, M. L. Sim, and F. W.-L. Kung, “Performance enhancement of
outdoor visible-light communication system using selective combining
receiver,” IET optoelectronics, vol. 3, no. 1, pp. 30–39, 2009.
[69] Z. Wang, J. Li, Q. Guo, and X. Gu, “Analysis on connectivity of inter-
orbit-links in a MEO/LEO double-layer satellite network,” Chinese J.
of Aeronautics, vol. 19, no. 4, p. 340, 2006.
[70] K. Kimura, K. Inagaki, and Y. Karasawa, “Double-layered inclined
orbit constellation for advanced satellite communications network,”
IEICE Trans. Commun., vol. 80, no. 1, pp. 93–102, 1997.
[71] J.-W. Lee, J.-W. Lee, T.-W. Kim, and D.-U. Kim, “Satellite over
satellite (SOS) network: A novel concept of hierarchical architecture
and routing in satellite network,” in Proc. 25th Annual IEEE Conf. on
Local Computer Netw., (LCN), 2000, pp. 392–399.
[72] T. Jono, Y. Takayama, N. Kura et al., “OICETS on-orbit laser
communication experiments,” in Free-Space Laser Communication
Technologies XVIII, vol. 6105, 2006, p. 610503.
[73] Y. Fujiwara, M. Mokuno, T. Jono, T. Yamawaki et al., “Optical
inter-orbit communications engineering test satellite (oicets),” Acta
Astronautica, vol. 61, no. 1-6, pp. 163–175, 2007.
[74] S. Yamakawa, T. Hanada, H. Kohata, and Y. Fujiwara, “JAXA’s efforts
toward next generation space data-relay satellite using optical inter-
orbit communication technology,” in Free-Space Laser Communication
Technologies XXII, vol. 7587, 2010, p. 75870P.
[75] T. Jono, Y. Takayama, K. Shiratama, I. Mase et al., “Overview of the
inter-orbit and the orbit-to-ground laser communication demonstration
by OICETS,” in Free-Space Laser Communication Technologies XIX
and Atmospheric Propagation of Electromagnetic Waves, vol. 6457.
Int. Society for Optics and Photonics, 2007, p. 645702.
[76] ESA, “Laser link offers high-speed delivery,” http://www.esa.int/
Applications/Observing the Earth/Copernicus/Sentinel-1/Laser link
offers high-speed delivery, online; accessed 14 Sep. 2020.
[77] D. Trondle, P. M. Pimentel, C. Rochow et al., “Alphasat-Sentinel-
1A optical inter-satellite links: run-up for the European data relay
satellite system,” in Free-Space Laser Communication and Atmospheric
Propagation XXVIII, vol. 9739, 2016, pp. 1 – 6.
[78] A. Grover, A. Sheetal, and V. Dhasarathan, “20 Gbit/s-40 GHz OFDM
based LEO-GEO radio over inter-satellite optical wireless communica-
tion (Ro-IsOWC) system using 4-QAM modulation,” Optik, vol. 206,
p. 164295, 2020.
[79] S. Li, D. Grace, J. Wei, and D. Ma, “A novel guaranteed handover
scheme for HAP communications systems with adaptive modulation
and coding,” in IEEE 72nd Veh. Techno. Conf. (VTC). IEEE, 2010,
pp. 1–5.
[80] Y. Zhou, F. Sun, and B. Zhang, “A novel qos routing protocol for leo
and meo satellite networks,” Int. J. of Satellite Commun. and Netw.,
vol. 25, no. 6, pp. 603–617, 2007.
[81] H. Yan, J. Guo, X. Wang, Y. Zhang, and Y. Sun, “Topology analysis
of inter-layer links for LEO/MEO double-layered satellite networks,”
in Space Information Networks, Q. Yu, Ed., 2018, pp. 145–158.
[82] G. Jull, “SHARP (stationary high altitude relay platform) telecommu-
nications missions and systems,” GTC, vol. 2, pp. 955–959, 1985.
[83] AeroVironment, Inc., “Aerovironment announces joint venture and
solar high-altitude long-endurance unmanned aircraft system develop-
ment program,” https://www.avinc.com/resources/press-releases/view/
solar-high-altitude-long-endurance-uas, online; accessed 14 Sep. 2020.
[84] S. Karapantazis and F. Pavlidou, “Broadband communications via high-
altitude platforms: A survey,” IEEE Commun. Surveys Tuts., vol. 7,
no. 1, pp. 2–31, 2005.
[85] F. A. d’Oliveira, F. C. L. d. Melo, and T. C. Devezas, “High-altitude
platforms—present situation and technology trends,” J. of Aerospace
Techno. and Management, vol. 8, no. 3, pp. 249–262, 2016.
[86] A. Aragon-Zavala, J. L. Cuevas-Ru´
ız, and J. A. Delgado-Pen´
ın, High-
altitude platforms for wireless communications. Wiley Online Library,
2008, vol. 5.
[87] CAPANINA, “CAPANINA test results summary report,” https:
//www.capanina.org/documents/CAP-D22a-WP44-CGS-PUB-01.pdf,
online; accessed 14 Sep. 2020.
16
[88] Stratxx lnc., “Stratxx,” http://www.stratxx.com/cgi-sys/
suspendedpage.cgi, online; accessed 14 Sep. 2020.
[89] Elevate, “Elevate,” http://www.zero2infinity.space/elevate/, online; ac-
cessed 14 Sep. 2020.
[90] Google, “Connecting people everywhere,” https://www.loon.com, on-
line; accessed 14 Sep. 2020.
[91] Airbus lnc., “Zephyr pioneering the stratosphere,” https:
//www.airbus.com/defence/uav/zephyr.html, online; accessed 14
Sep. 2020.
[92] A. Cox, “Flying aquila: Early lessons from the first full-
scale test flight and the path ahead - facebook engineering,”
https://engineering.fb.com/connectivity/flying-aquila-early-lessons-
from-the- first-full- scale-test- flight-and- the-path- ahead/, online;
accessed 14 Sep. 2020.
[93] Thales, “What’s up with stratobus?” https://www.thalesgroup.com/en/
worldwide/space/news/whats-stratobus, online; accessed 14 Sep. 2020.
[94] HAPSMobile lnc., “Hapsmobile,” https://www.hapsmobile.com/en/,
online; accessed 14 Sep. 2020.
[95] Prismatic lnc., “Phasa-35,” http://prismaticltd.co.uk/products/phasa-
35/, online; accessed 14 Sep. 2020.
[96] T. Tozer and D. Grace, “High-altitude platforms for wireless communi-
cations,” Electronics & Commun. Eng. J., vol. 13, no. 3, pp. 127–137,
2001.
[97] A. Mohammed, A. Mehmood, F.-N. Pavlidou, and M. Mohorcic,
“The role of high-altitude platforms (HAPs) in the global wireless
connectivity,” Proc. of the IEEE, vol. 99, no. 11, pp. 1939–1953, 2011.
[98] D. Grace and M. Mohorcic, Broadband communications via high
altitude platforms. John Wiley and Sons, 2011.
[99] D. Grace, M. H. Capstick, M. Mohorcic, J. Horwath, M. B. Pallavicini,
and M. Fitch, “Integrating users into the wider broadband network via
high altitude platforms,” IEEE Wireless Commun., vol. 12, no. 5, pp.
98–105, 2005.
[100] J. Thornton, A. D. White, and T. C. Tozer, “A WiMAX payload for
high altitude platform experimental trials,” EURASIP J. on Wireless
Commun. and Netw., vol. 2008, pp. 1–9, 2008.
[101] Y. Yang, R. Zong, X. Gao, and J. Cao, “Channel modeling for high-
altitude platform: A review,” in Int. Symp. on Intelligent Signal Process.
and Commun. Sys., 2010, pp. 1–4.
[102] A. Nauman and M. Maqsood, “System design and performance evalu-
ation of high altitude platform: Link budget and power budget,” in 19th
Int. Conf. on Adv. Commun. Techno., (ICACT), 2017, pp. 138–142.
[103] L. F. Abdulrazak, “Stratospheric winds and rain effect on haps backhaul
link performance,” Kurdistan J. of App. Res., vol. 2, no. 3, pp. 252–259,
2017.
[104] Y. Shibata, N. Kanazawa, M. Konishi, K. Hoshino, Y. Ohta, and
A. Nagate, “System design of Gigabit HAPS mobile communications,”
IEEE Access, vol. 8, pp. 157 995–158 007, 2020.
[105] J. Zhao, Q. Wang, Y. Li, J. Zhou, and W. Zhou, “Ka-band based channel
modeling and analysis in high altitude platform (HAP) system,” in
IEEE 91st Vehicular Techno. Conf., (VTC), 2020, pp. 1–5.
[106] K. Popoola, D. Grace, and T. Clarke, “Capacity and coverage analysis
of high altitude platform (HAP) antenna arrays for rural vehicular
broadband services,” in IEEE 91st Vehicular Techno. Conf., (VTC),
2020, pp. 1–5.
[107] J. Lun, D. Grace, N. Morozs, P. D. Mitchell, Y. Chu, A. Fisusi,
O. Awe, and R. E. Sheriff, “Tv white space broadband for rural
communities using solar powered high altitude platform and terrestrial
infrastructures,” White Paper, 2017.
[108] M. Alzenad, M. Z. Shakir, H. Yanikomeroglu, and M.-S. Alouini,
“FSO-based vertical backhaul/fronthaul framework for 5G+ wireless
networks,” IEEE Communications Mag., vol. 56, no. 1, pp. 218–224,
2018.
[109] E. Katimertzoglou, D. Vouyioukas, P. Veltsistas, and P. Constantinou,
“Optical interplatform links scenarios for 20 km altitude,” in 16th IST
Mobile and Wireless Commun. Summit, 2007, pp. 1–5.
[110] H. Akbar et al., “Ber performance analysis of apd-based fso system
for optical inter-haps link,” in 1st Int. Conf. on Wireless and Telematics
(ICWT), 2015, pp. 1–5.
[111] A. A. Farid and S. Hranilovic, “Outage capacity optimization for free-
space optical links with pointing errors,” J. of Lightwave technology,
vol. 25, no. 7, pp. 1702–1710, 2007.
[112] W. Gappmair, S. Hranilovic, and E. Leitgeb, “OOK performance for
terrestrial FSO links in turbulent atmosphere with pointing errors
modeled by hoyt distributions,” IEEE Commun. Lett., vol. 15, no. 8,
pp. 875–877, 2011.
[113] F. Yang, J. Cheng, and T. A. Tsiftsis, “Free-space optical communi-
cation with nonzero boresight pointing errors,” IEEE Trans. Commun.,
vol. 62, no. 2, pp. 713–725, 2014.
[114] H. AlQuwaiee, H. Yang, and M. Alouini, “On the asymptotic capacity
of dual-aperture FSO systems with generalized pointing error model,”
IEEE Trans. Wireless Commun., vol. 15, no. 9, pp. 6502–6512, 2016.
[115] H. Kaushal and G. Kaddoum, “Optical communication in space:
Challenges and mitigation techniques,” IEEE Commun. Surveys Tuts.,
vol. 19, no. 1, pp. 57–96, 2017.
[116] A. Habash, L. C. Andrews, and R. L. Phillips, “Mathematical model for
the irradiance probability density function of a laser beam propagating
through turbulent media,” Optical Engineering, vol. 40, no. 8, pp.
1554–1562, 2001.
[117] A. K. Majumdar, “Free-space laser communication performance in the
atmospheric channel,” J. of Optical and Fiber Commun. Rep., vol. 2,
no. 4, pp. 345–396, 2005.
[118] A. Trichili, M. A. Cox, B. S. Ooi, and M.-S. Alouini, “Roadmap to
free space optics,” J. Opt. Soc. Am. B, vol. 37, no. 11, pp. 184–201,
Nov. 2020.
[119] M. Guan, Z. Wu, Y. Cui, and M. Yang, “Channel modeling and char-
acteristics for high altitude platform stations communication system,”
J. of Internet Techno., vol. 21, no. 3, pp. 891–897, 2020.
[120] I. R. Palma-L´
azgare and J. A. Delgado-Pen´
ın, “WiMAX haps-
based downlink performance employing geometrical and statistical
propagation-channel characteristics,” URSI Radio Science Bulletin, vol.
2010, no. 333, pp. 50–66, 2010.
[121] I. Zakia, “Capacity of HAP-MIMO channels for high-speed train
communications,” in 3rd Int. Conf. on Wireless and Telematics (ICWT),
2017, pp. 26–30.
[122] J. Horwath, N. Perlot, M. Knapek, and F. Moll, “Experimental veri-
fication of optical backhaul links for high-altitude platform networks:
Atmospheric turbulence and downlink availability,” Int. J. of Satellite
Commun. and Netw., vol. 25, no. 5, pp. 501–528, 2007.
[123] E. T. Michailidis and A. G. Kanatas, “Three-dimensional HAP-MIMO
channels: Modeling and analysis of space-time correlation,” IEEE
Trans. on Veh. Techno., vol. 59, no. 5, pp. 2232–2242, 2010.
[124] P. Sudheesh, N. Sharma, M. Magarini, and P. Muthuchidambaranathan,
“Effect of imperfect CSI on interference alignment in multiple-high
altitude platforms based communication,” Physical Commun., vol. 29,
pp. 336–342, 2018.
[125] J. Thornton, D. Grace, M. H. Capstick, and T. C. Tozer, “Optimizing an
array of antennas for cellular coverage from a high altitude platform,”
IEEE Trans. on Wireless Commun., vol. 2, no. 3, pp. 484–492, 2003.
[126] D. Xu, X. Yi, Z. Chen, C. Li, C. Zhang, and B. Xia, “Coverage
ratio optimization for HAP communications,” in IEEE 28th Annual
Int. Symp. on Pers., Indoor, and Mobile Radio Commun., (PIMRC),
2017, pp. 1–5.
[127] C. K. Datsikas, K. P. Peppas, N. C. Sagias, and G. S. Tombras, “Se-
rial free-space optical relaying communications over gamma-gamma
atmospheric turbulence channels,” IEEE/OSA J. of Optical Commun.
and Netw., vol. 2, no. 8, pp. 576–586, 2010.
[128] M. Sharma, D. Chadha, and V. Chandra, “High-altitude platform
for free-space optical communication: Performance evaluation and
reliability analysis,” IEEE/OSA J. of Optical Commun. and Netw.,
vol. 8, no. 8, pp. 600–609, 2016.
[129] E. T. Michailidis, N. Nomikos, P. Bithas, D. Vouyioukas, and A. G.
Kanatas, “Outage probability of triple-hop mixed RF/FSO/RF strato-
spheric communication systems,” in IEEE 10th Int. Conf. on Adv. in
Satellite and Space Commun., (SPACOMM), 2018, pp. 1–6.
[130] W. L. Lim, Y. C. Foo, and R. Tafazolli, “Adaptive softer handover
algorithm for high altitude platform station UMTS with onboard
power resource sharing,” in The 5th Int. Symp. on Wireless Personal
Multimedia Commun., vol. 1, 2002, pp. 52–56.
[131] K. Katzis and D. Grace, “Inter-high-altitude-platform handoff for
communications systemswith directional antennas,” URSI Radio Sci.
Bulletin, vol. 2010, no. 333, pp. 29–38, 2010.
[132] S. Li, D. Grace, J. Wei, and D. Ma, “Directional traffic-aware intra-
HAP handoff scheme for HAP communications systems,” URSI Radio
Sci. Bulletin, vol. 2010, no. 334, pp. 11–18, 2010.
[133] S. Li, L. Wang, G. David, and D. Ma, “Cooperative directional inter-cell
handover scheme in high altitude platform communications systems,”
J. of Electronics (China), vol. 28, no. 2, p. 249, 2011.
[134] S. Ni, S. Jin, and H. Hong, “A handover decision algorithm with an
adaptive threshold applied in HAPS communication system,” in Theory,
Methodology, Tools and Applications for Modeling and Simulation of
Complex Systems, 2016, pp. 38–47.
17
[135] P. Seitzer, “Mega-constellations of satellites and optical astronomy,”
AAS, pp. 410–430, 2020.
[136] P. Pace and G. Aloi, “Satellite-HAP network supporting multilayered
QoS routing in the sky,” IETE J. of Res., vol. 56, no. 3, pp. 163–174,
2010.
[137] G. Pan, J. Ye, Y. Tian, and M.-S. Alouini, “On HARQ schemes in
satellite-terrestrial transmissions,” IEEE Trans. Wireless Commun., pp.
1–1, 2020.
[138] J. Ye, S. Dang, B. Shihada, and M.-S. Alouini, “Space-air-ground
integrated network: Outage performance analysis,” IEEE Trans. on
Wireless Commun., pp. 1–1, 2020.
[139] N. Zhang, S. Zhang, P. Yang, O. Alhussein, W. Zhuang, and X. S.
Shen, “Software defined space-air-ground integrated vehicular net-
works: Challenges and solutions,” IEEE Commun. Mag., vol. 55, no. 7,
pp. 101–109, 2017.
[140] Z. Zhou, J. Feng, C. Zhang, Z. Chang, Y. Zhang, and K. M. S. Huq,
“SAGECELL: Software-defined space-air-ground integrated moving
cells,” IEEE Commun. Mag., vol. 56, no. 8, pp. 92–99, 2018.
[141] B. Di, H. Zhang, L. Song, Y. Li, and G. Y. Li, “Ultra-dense LEO:
Integrating terrestrial-satellite networks into 5G and beyond for data
offloading,” IEEE Trans. on Wireless Commun., vol. 18, no. 1, pp.
47–62, 2019.
[142] X. Zhu, C. Jiang, L. Kuang, N. Ge, S. Guo, and J. Lu, “Cooperative
transmission in integrated terrestrial-satellite networks,” IEEE Network,
vol. 33, no. 3, pp. 204–210, 2019.
[143] H. Yao, L. Wang, X. Wang, Z. Lu, and Y. Liu, “The space-terrestrial
integrated network: An overview,” IEEE Commun. Mag., vol. 56, no. 9,
pp. 178–185, 2018.
[144] A. Vanelli-Coralli, G. E. Corazza, M. Luglio, and S. Cioni, “The
ISICOM architecture,” in Int. Works. on Satellite and Space Commun.,
2009, pp. 104–108.
[145] J. Pulliam, Y. Zambre, A. Karmarkar, V. Mehta, J. Touch, J. Haines, and
M. Everett, “TSAT network architecture,” in IEEE Military Commun.
Conf., (MILCOM), 2008, pp. 1–7.
[146] N. Perlot, E. Duca, J. Horwath, D. Giggenbach, and E. Leitgeb,
“System requirements for optical hap-satellite links,” in 6th Int. Symp.
on Commun. Sys., Netw. and Digital Signal Process., 2008, pp. 72–76.
[147] M. Q. Vu, N. T. Nguyen, H. T. Pham, and N. T. Dang, “Performance
enhancement of LEO-to-ground FSO systems using all-optical HAP-
based relaying,” Physical Commun., vol. 31, pp. 218–229, 2018.
[148] A. Zappone, M. Debbah, and Z. Altman, “Online energy-efficient
power control in wireless networks by deep neural networks,” in
IEEE 19th Int. Works. on Signal Process. Adv. in Wireless Commun.,
(SPAWC), 2018, pp. 1–5.
[149] A. Asheralieva, “Bayesian reinforcement learning-based coalition for-
mation for distributed resource sharing by device-to-device users in
heterogeneous cellular networks,” IEEE Trans. Wireless Commun.,
vol. 16, no. 8, pp. 5016–5032, 2017.
[150] S. S. Mwanje, L. C. Schmelz, and A. Mitschele-Thiel, “Cognitive cel-
lular networks: A Q-learning framework for self-organizing networks,”
IEEE Trans. Netw. Service Manag., vol. 13, no. 1, pp. 85–98, 2016.
[151] N. Kato, Z. M. Fadlullah, F. Tang, B. Mao, S. Tani, A. Okamura, and
J. Liu, “Optimizing space-air-ground integrated networks by artificial
intelligence,” IEEE Wireless Commun., vol. 26, no. 4, pp. 140–147,
2019.
[152] S. Zhou, G. Wang, S. Zhang, Z. Niu, and X. S. Shen, “Bidirectional
mission offloading for agile space-air-ground integrated networks,”
IEEE Wireless Commun., vol. 26, no. 2, pp. 38–45, 2019.
[153] W. Abderrahim, O. Amin, M. Alouini, and B. Shihada, “Latency-aware
offloading in integrated satellite terrestrial networks,” IEEE Open J. of
the Commun. Soc., vol. 1, pp. 490–500, 2020.
[154] C. Niephaus, J. M ¨
odeker, and G. Ghinea, “Toward traffic offload in
converged satellite and terrestrial networks,” IEEE Trans. Broadcast.,
vol. 65, no. 2, pp. 340–346, 2019.
[155] R. Alghamdi, R. Alhadrami, D. Alhothali et al., “Intelligent surfaces
for 6G wireless networks: A survey of optimization and performance
analysis techniques,” IEEE Access, pp. 1–15, 2020.
[156] S. Alfattani, W. Jaafar, Y. Hmamouche et al., “Aerial platforms with
reconfigurable smart surfaces for 5G and beyond,” 2020. [Online].
Available: https://arxiv.org/abs/2006.09328
[157] K. Tekbıyık, G. K. Kurt, A. R. Ekti, A. G¨
orc¸in, and
H. Yanikomeroglu, “Reconfigurable intelligent surface empowered
terahertz communication for LEO satellite networks,” arXiv preprint,
2020. [Online]. Available: https://arxiv.org/abs/2007.04281
[158] S. Alfattani, W. Jaafar, Y. Hmamouche, H. Yanikomeroglu, and
A. Yongac¸oglu, “Link budget analysis for reconfigurable smart
surfaces in aerial platforms,” 2020. [Online]. Available: https:
//arxiv.org/abs/2008.12334
[159] Y. Yang, M. Xu, D. Wang, and Y. Wang, “Towards energy-efficient
routing in satellite networks,” IEEE J. Sel. Areas Commun., vol. 34,
no. 12, pp. 3869–3886, 2016.
[160] J. Zhang, X. Zhang, M. A. Imran, B. Evans, Y. Zhang, and W. Wang,
“Energy efficient hybrid satellite terrestrial 5G networks with software
defined features,” J. of Commun. and Netw., vol. 19, no. 2, pp. 147–161,
2017.
[161] A. Rahmati, Y. Yapici, N. Rupasinghe, I. Guvenc, H. Dai, and
A. Bhuyan, “Energy efficiency of RSMA and NOMA in cellular-
connected mmwave UAV networks,” in IEEE Int. Conf. on Commun.
Works., (ICC Workshops), 2019, pp. 1–6.
[162] W. Jaafar, S. Naser, S. Muhaidat et al., “Multiple access in aerial
networks: From orthogonal and non-orthogonal to rate-splitting,”
2020. [Online]. Available: https://arxiv.org/abs/2005.13122
[163] H. E. Hammouti, D. Hamza, B. Shihada, M.-S. Alouini, and J. S.
Shamma, “The optimal and the greedy: Drone association and
positioning schemes for internet of UAVs,” 2020. [Online]. Available:
https://arxiv.org/abs/2004.00839
