SPECIAL SECTION ON INTERNET OF SPACE: NETWORKING ARCHITECTURES

AND PROTOCOLS TO SUPPORT SPACE-BASED INTERNET SERVICES

Received December 8, 2021, accepted January 7, 2022, date of publication January 25, 2022, date of current version February 1, 2022.
Digital Object Identifier 10.1109/ACCESS.2022.3146021

SALSA: A Scheduling Algorithm for

LoRa to LEO Satellites
MOHAMMAD AFHAMISIS 1,2 , (Member, IEEE), AND
MARIA RITA PALATTELLA 1 , (Member, IEEE)
1 Luxembourg Institute of Science and Technology, 4362 Esch-sur-Alzette, Luxembourg
2 Faculty of Science, Technology and Medicine, University of Luxembourg, 4365 Esch-sur-Alzette, Luxembourg
Corresponding author: Mohammad Afhamisis ([email protected])
This work was supported by the Design of LoRaWAN protocol optimisation over SATellite Connection for precision agriculture
applications (LORSAT) Project, through the National Research Fund Luxembourg (FNR), under Grant CORE/C19/IS/13705191.

ABSTRACT Satellites are going to revolutionize the Internet of Things (IoT) connectivity enabling the
ubiquitous coverage targeted by 5G and upcoming 6G networks. While the potentials of satellite IoT are
unquestionable, many challenges still remain unsolved. In the present work, we design a Long Range Wide
Area Network (LoRaWAN) with mobile gateways (GWs) installed on Low Earth Orbit (LEO) satellites.
We address synchronization issues due to intermittent link availability between the End Devices (EDs)
and the GW, and we propose a Scheduling Algorithm for LoRa to LEO Satellites (SALSA). SALSA
ensures reliable communication, avoiding packets drops and packet collisions, by using a Time Division
Multiple Access (TDMA) approach, rather than classic ALOHA-based LoRa. The uplink transmissions of
the EDs are scheduled considering the satellite availability period, the satellite footprint, and the specific
visit time for each ED. Two different policies are proposed: a First Come First Served (FCFS), and a FAIR
policy. Simulation results, obtained in real conditions, with real satellites visibility, demonstrate the better
performance achievable with SALSA (regardless of the policy) compared to standard LoRa. The FAIR policy
outperforms the FCFS policy by giving all the EDs an equal chance to transmit, even to those that are visited
last by the satellite. The performance of SALSA can be further improved considering larger constellations
of LEO satellites.

INDEX TERMS Satellite communication, Internet of Things, performance evaluation, optimization meth-
ods, scheduling algorithm, LoRaWAN, synchronization, fairness.

I. INTRODUCTION reduce latency to very low levels, compared to GEO satellites,

Satellites are going to revolutionize Internet of Things (IoT) it will still not be suitable to meet the needs of 5G Ultra
connectivity enabling the ubiquitous coverage targeted by Reliable Low Latency Communications (URLLC), such as
5G and Beyond 5G networks. Satellite for IoT is becoming Augmented Reality/Virtual Reality (AR/VR) applications,
more and more affordable, available, and accessible, thanks and Tactile Internet. However, the delay is more than ade-
to the launch of several low-cost miniaturized Low Earth quate for many other services that do not require extremely
Orbit (LEO) satellites (CubeSats) [1]. Those LEO satellites low latency, like precision agriculture, maritime surveillance,
are the most appealing ones for IoT applications, due to the environmental monitoring, etc.
shorter delay that they introduce (' 40 ms), compared to Geo- Moreover, being closer to the Earth, LEO satellites can sup-
stationary Earth Orbit (GEO) satellites (' 500 ms). However, port direct communication with IoT devices on the ground.
their intrinsic orbital properties imply limited visibility time. Over the last years, several satellite operators and IoT compa-
This issue can be overcome by using large constellations of nies have recognized the potential innovation and consequent
LEO satellites, able to provide almost continuous coverage; business opportunities in enabling direct IoT to satellite.
and it will be further solved with relay networks from LEO Among different IoT technologies, the Long-Range Wide
to GEO satellites. While LEO satellite constellations can Area Network (LoRaWAN) technology [2] has played, and
is still having a significant role in the real deployment of
The associate editor coordinating the review of this manuscript and IoT. Companies like Lacuna Space [3] and Wyld [4] have
approving it for publication was Rute C. Sofia . already demonstrated how global IoT connectivity can be
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
11608 VOLUME 10, 2022
M. Afhamisis, M. R. Palattella: SALSA: Scheduling Algorithm for LoRa to LEO Satellites

achieved by integrating a LoRaWAN gateway on a LEO The remainder of this paper is structured as follows.
satellite. Section II introduces the system architecture and the back-
Satellite LoRaWAN has drawn the attention also of ground information about the LoRa PHY and LoRaWAN
academia, and researchers. Several works have identified the MAC protocols. In Section III, we describe the proposed
open research challenges and proposed a preliminary net- SALSA algorithm in all the details, including the system
work architecture [5]–[8]. In a recent survey [5], the authors assumptions and the scheduling policies. Section IV first
have discussed the pros and cons of satellite access for IoT, presents the simulated environment and then, the perfor-
current limitations, and ongoing initiatives for making it mance achievable with SALSA, considering real conditions.
possible. Very few works have provided technical solutions, Through simulation results, we show that SALSA cancels by
as enhancement of LoRa PHY modulation [9], and manage- design collisions due to random ALOHA MAC, and packet
ment of Doppler effect [10]. drops due to intermittent link availability. When using the
While the use of a satellite gateway increases the cov- FAIR policy and having a constellation of satellites, the end
erage range of the LoRaWAN network, it also introduces devices get an equal chance of transmitting several uplink
several challenges due to its limited visibility time. The prob- packets. Finally, Sec. V draws the conclusion and future
ability of collision due to the random ALOHA-like access research direction.
scheme - already high in classical terrestrial LoRaWAN
networks [11], [12] - can only get worst since many End II. SYSTEM ARCHITECTURE
Devices (EDs) will compete for the same resource and We consider a satellite LoRaWAN network, whose architec-
attempt an uplink transmission during the limited availabil- ture is depicted in Fig.1. Several end-devices (EDs) deployed
ity of the gateway. To address such a problem and ensure on the ground can communicate through wireless LoRa links
reliable communication, scheduling techniques are needed. with a LoRaWAN gateway (GW) installed on a LEO satellite.
Several ones have been proposed in literature for terres- Only the EDs within the satellite footprint can reach the GW.
trial LoRaWAN networks [13]–[20]. Zorbas et al. tackled
the issue of a LoRaWAN mobile gateway, installed on a
drone [21], [22]. To deal with the limited availability of
the flying gateway, they have proposed to transmit a bulk
of data and reduce the ToA for the grouped transmissions.
A couple of recent works considered direct LoRa to satellite,
but they did not enhance network performance with schedul-
ing techniques. Ullah et al. [23] modeled the PHY channel
between the EDs and the satellite gateway and assumed
EDs transmit according to the ALOHA-like LoRa standard.
Also, Tondo et al. [24] proposed a system model for direct
IoT to satellite, with a constellation of LEO satellites. They
evaluated throughput and packet loss rate as a function of
FIGURE 1. LoRaSAT Network Architecture.
the erasure probabilities at each satellite. As in our work,
they demonstrated the network performance could improve
with larger constellations. Besides that, they assumed the EDs The footprint represents the area of the Earth covered by
transmit using slotted ALOHA. microwave radiation from a satellite dish (transponder). Thus,
In this paper for the first time, we address reliability and it corresponds to the coverage area. The size of the footprint
scalability issues in LoRa to LEO satellites communication. depends on several factors: the location of the satellite along
We propose SALSA, a scheduling algorithm that cancels its orbit, the shape, and size of the beam produced by its
packet drops and packet collisions by design, and ensures fair transponder, and the distance from the earth. In particular,
access to all the nodes in the network. We have designed the the footprint is impacted by the elevation angle θe , which
algorithm considering the worst-case scenario (i.e., a single is the angle between the beam pointing direction, directly
satellite available), while it is still applicable and extend- towards the satellite, and the local horizontal plane. As shown
able to LEO constellations. Two different access policies are in Fig.2, the satellite footprint decreases with increasing θe .
proposed: a First Come First Served (FCFS), and a FAIR In the example, the size of the footprint A is smaller than
policy. Simulation results, obtained in real conditions, with the size of footprint B, being θeA > θeB . The optimum
real satellites visibility, demonstrate the better performance value of θe depends on the satellite orbit, the budget link,
achievable with SALSA (regardless of the policy) compared and the communication link design. To properly schedule
to standard LoRa. The FAIR policy outperforms the FCFS uplink transmission from the EDs to the GW, it is of foremost
policy by giving all the EDs an equal chance to transmit, even importance to know the movement of the satellite and its
to those that are visited last by the satellite. The performance relative position to the IoT devices. Only with such infor-
of SALSA can be further improved considering larger con- mation, it is possible ensuring precise synchronization with a
stellations of LEO satellites. fixed θe .

VOLUME 10, 2022 11609

 M. Afhamisis, M. R. Palattella: SALSA: Scheduling Algorithm for LoRa to LEO Satellites

by Semtech and not available in Commercial off-the-shelf

(COTS) devices. Therefore, in our work, aiming to provide
a concrete and feasible solution, we assumed EDs communi-
cate over LoRa with SF = 12. Network scalability of LoRa
to LEO satellites could be further improved in future work,
combining SALSA with LR-FHSS. In fact, by leveraging
on random frequency hopping and dividing the channel into
several sub-channels, LR-FHSS allows using several data
rates at the same time (two in the USA FCC and four in the
EU ETSI) [9]. Many more transmissions could be scheduled
in parallel during the limited satellite visibility, on orthogonal
FIGURE 2. Relation between the elevation angle and the satellite channels, without having collisions.
footprint. The footprint A is smaller than the footprint B in size being the
elevation angle θeA > θeB .
B. MAC LAYER
At the Medium Access Control (MAC) layer, the network
While we describe the architecture with a single satellite adopts the LoRaWAN protocol, based on pure ALOHA ran-
and thus a single GW, the system can be extendible to multiple dom access [26]. The standard [2] defines three main channel
gateways installed on a constellation of LEO satellites. access strategies. Class A (Aloha) includes the basic set of
Multiple GWs are connected to a central LoRaWAN Net- features needed by any ED to join a LoRaWAN network.
work Server (LNS) on the ground through a reliable satellite Class A EDs use ALOHA protocol for an uplink packet
backhauling network (via a ground satellite gateway). The transmission, followed by two short downlink receive win-
GW acts as a forwarding device. It is responsible only for dows, RX1 and RX2 . Thus, the downlink communication is
re-dispatching raw data packets from EDs toward the LNS, triggered by the ED. Each downlink frame needs to wait
encapsulating them in IP packets. The LNS manages the for the uplink communication. The LoRaWAN specification
GWs and the entire end-to-end system; it is responsible for provides two enhancements to Class A: Class B (Beacon)
the LoRaWAN protocol configuration and execution in the and Class C (Continuous) to increase the downlink possibil-
network; it collects and decodes the received packets. In the ities. These downlink protocol enhancements are paid with
present work, we focus on the first segment of the end-to- additional energy consumption. Class B EDs open additional
end system, i.e., the intermittent link between the EDs and receive windows at scheduled times. The GW transmits bea-
the LEO satellite(s). We assume there is a ground satellite cons to keep the synchronization with EDs and opens new
gateway in the visibility of the satellite, and it ensures the downlink slots. Thus, the server can directly deliver frames
reception of the packet at the LNS on the ground. to correspondent EDs without waiting for them to start an
uplink communication. Class C EDs keep their radio modules
A. PHY LAYER always on in reception mode, and thus they are available
At the Physical (PHY) layer, the communication between all the time for listening to downlink traffic unless they are
EDs and GWs follows the Long Range (LoRa) transmitting. In our work, we assumed EDs operate in Class
technology [25] based on a spread spectrum modulation. A and generate uplink traffic, being this the operative mode
The LoRa modulation is available with several Spreading supported by all the COTS devices. In the future, we aim to
Factors (SFs) to trade-off between the resulting bit rate, design scheduling algorithms for bidirectional communica-
the transmission range, and the energy required to transmit. tion, and thus, we will also consider Class B and Class C
According to the standard, the SF can assume integer val- devices, better supporting downlink traffic.
ues between 7 and 12. A lower value of SF allows faster
modulation, shorter Time-on-Air (ToA), and lower energy III. SALSA SCHEDULING ALGORITHM
consumption. On the other side, a signal with lower SF is In the present work, we propose SALSA, a Scheduling Algo-
more affected by noise, attenuation, and interference. There- rithm for LoRa to LEO SAtellites. In this Section, we describe
fore, higher values of SF are more suitable for long-range the algorithm in all its details, from the system assumptions,
communication. Recently, a new PHY layer, Long Range to the different access policies.
Frequency Hopping Spread Spectrum (LR-FHSS) [9] was
proposed to address extremely long-range and large-scale A. SYSTEM ASSUMPTIONS
communication scenarios such as direct IoT satellite. At the In the design of the algorithm, we assume that the LNS is
core of LR-FHSS, there is a fast frequency hopping technique aware of the location of the EDs and the availability time of
that allows higher network capacity while offering the same the gateway (i.e., the visibility period of the satellite). The
radio link budget as LoRa. The LR-FHSS mechanism is EDs may have been pre-registered in the network, using Over
one of the few concrete attempts to optimize the LoRa The Air (OTA) registration while the gateway was still on the
PHY protocol for direct communication with LEO satellites. ground. All the EDs operate in Class A. Moreover, we assume
Currently, the LR-FHSS is still under standardization process that the LNS sends in unicast (during the RX windows) the

11610 VOLUME 10, 2022

M. Afhamisis, M. R. Palattella: SALSA: Scheduling Algorithm for LoRa to LEO Satellites

scheduling table to each ED in the network. To consider the can have variable time duration: in the simpler case, a slide is
deviation of the current orbit of the satellite from its initial equal to the time needed for completing a single uplink trans-
orbit, periodic scheduling updates should be sent directly to mission Tr . It is the case when there is no overlap between
each device using unicast messages. Note that the exchange concurrent transmissions from several EDs because the EDs
of the configuration parameters and scheduling tables is out are one far away from another. On the contrary, when there is
of the scope of the present work, the reason why we made an overlap (i.e., a dense area with many EDs concentrated
such assumptions. together), and an ED cannot finish a transmission before
The LNS is in charge of scheduling the uplink transmis- another ED enters the satellite coverage, the LNS allocates
sions for each ED and assigning priorities. To guarantee long- a larger slide. The slide’s duration is set to Ns × Tr , where
range transmission from the EDs to the satellite gateway, Ns is the number of neighbor nodes within the identified
we assume that all the EDs transmit using SF = 12. Using dense area. The LNS defines the starting time TSi of the
a single value of the SF implies that only a transmission per EDi transmission. If the Start of the Satellite Visibility for
time can be scheduled, without incurring a collision. Accord- that EDi , TSSVi overlaps with an ongoing transmission from
ing to the standard, a single uplink transmission keeps the another EDj , the EDi has to wait till TEj (ending time of EDj
channel busy for a Time on Air (ToA) period. We propose to transmission). Either wise, it can start transmitting as soon
add two Guard Time, Tg , one before and one after the ToA for as it sees the satellite (see Algorithm 1). In the example in
taking into account possible synchronization issues between Figure 4, the ED4 will not transmit as soon as it enters the
the ED and the mobile gateway (see Fig. 3). Thus, whenever satellite footprint, but after the ED3 finishes its transmission.
the LNS schedules a transmission for an ED, it reserves the By doing so, SALSA with its TDMA approach and First
channel for that ED for a period of time equal to Tr = Come First Served (FCFS) policy results to be a collision-
ToA + 2 × Tg . free scheduling algorithm. It intrinsically avoids collision, not
allowing conflicting parallel overlapping transmissions.

FIGURE 3. Time reserved Tr for each ED for an uplink transmission. 2 × Tg

have been introduced to mitigate uncertainties and inaccuracies of the
synchronisation between the ED and the satellite gateway. The receive
windows RX1 and RX2 are opened after the end of the uplink
transmission, according to the standard.

B. FCFS POLICY
The LNS builds the scheduling tables for each of the EDs in FIGURE 4. Scheduling of EDs transmissions according to the satellite
movement along its orbit, and the visit time of the EDs. EDs that are
the network based on their discovering time of the satellite. visited first have first chance to transmit.
The latter depends on the satellite footprint, the satellite
movement along its orbit, and the value of the elevation
angle. As shown in Fig.4, while the satellite moves along
Algorithm 1 SALSA With FCFS Policy
its orbit, the satellite footprint also moves in time: therefore,
different EDs will be progressively in the satellite visibility, for slidek do
one after the other, in different periods. To efficiently use the for EDi do
network resources and maximize the number of EDs that can if TSSVi < TEj then
communicate with the mobile gateway, the LNS gives priority TSi ← TEj F TX after last ED
to the EDs that are visited first by the satellite footprint. In the else
example in Fig.4, the EDi will transmit before the EDj being TSi ← TSSVi F TX when satellite is available
ri < rj , with ri relative distance of EDi from the satellite end if
footprint. end for
Many nodes may be concentrated in the same area end for
(i.e., dense area), one very close to another, and thus, while an
ED is still transmitting, another one could enter the satellite
footprint and attempt a transmission as well. To avoid such C. FAIR POLICY
a scenario generating collisions, we introduce the concept of The LEO satellite is available for a limited time, equal on
slide. We divide the satellite footprint into several slides. They average to 100s. The visibility time could be shorter than the

VOLUME 10, 2022 11611

 M. Afhamisis, M. R. Palattella: SALSA: Scheduling Algorithm for LoRa to LEO Satellites

FIGURE 5. Satellite footprint and slides. Cancellation of collision with TDMA approach and FCFS policy.

total time needed for scheduling at least one uplink transmis- D. OPTIMAL SCHEDULE
sion from all the EDs within every single slide. In a dense After scheduling the EDs’ transmissions according to the
area with hundred of EDs, several EDs that are visited last FAIR policy, there might be still several slots available,
by the satellite may never get a chance to transmit. They depending on the satellite visibility and the density of the EDs
will miss the satellite each time it is their turn to transmit, in the coverage area. Due to (i) the short availability time of
according to the FCFS policy. We propose the FAIR policy to the satellite and (ii) the LoRa duty cycle limitations imposed
overcome such shortcomings and ensure that all the EDs have by the LoRaWAN regional parameters, it is very unlucky
an equal chance to transmit. The logic behind it is described that a given ED have a second chance of transmitting within
by Algorithm 2. a single satellite visit. For instance, the maximum ToA for
{51 Bytes, EU868, SF12, 125KHz} [27] is equal to 2793.5ms
Algorithm 2 SALSA With Fair Policy while the duty cycle is 1% in the EU region. So the ED must
for slidek do wait for 279.35s before having a new uplink opportunity,
for EDi do while the availability time for a LEO satellite is shorter than
if ntxi ≤ ntxj then F Fairness check 120s at 500Km height above 30◦ elevation.
if TSSVi < TEj then While running the SALSA algorithm the LNS also imple-
TSi ← TEj F TX after last ED ments a check on the duty cycle limitations and further opti-
else mizes the use of the limited network resources. And whenever
TSi ← TSSVi F TX when satellite is available it identifies a gap (empty slot) in the schedule, it allocates it
end if to another EDi , at the conditions that: (1) The new expected
else reserved time, Tri , does not create any conflict in the schedul-
Do not TX F Give the chance to the next ED ing table with the already reserved slots for other EDs. (2) The
end if EDi has visibility with the satellite in that expected time.
end for
end for IV. PERFORMANCE EVALUATION
In this section we first describe the simulated environment
Let consider N EDs within a given slide of the satellite used for reproducing the system behaviour, and for imple-
footprint. The LNS first checks if the EDi had already several menting the SALSA algorithm. Then, we evaluate the per-
transmission opportunities, ntxi , higher than the other end formance achievable with SALSA, and compare them with
devices EDj , within the same slide. In this case, with ntxi > the benchmark solution (ALOHA).
ntxj , and j 6 = i, the EDi will not be granted an uplink slot.
On the contrary, if it can transmit because it transmitted fewer A. SIMULATED ENVIRONMENT
times than its neighbours, the LNS has to define the starting We have implemented the algorithm in a Matlab environ-
time TSi of its transmission, according to the FCFS policy (see ment to evaluate the performance achievable with SALSA,
Algorithm 1). FCFS, and FAIR policies. We consider network size variable

11612 VOLUME 10, 2022

M. Afhamisis, M. R. Palattella: SALSA: Scheduling Algorithm for LoRa to LEO Satellites

FIGURE 8. Packet drops during one month when the EDs transmit with
ALOHA-based LoRa, generating periodic traffic, every 30 minutes.

FIGURE 6. Location of EDs over the country of Luxembourg, following a

random uniform distribution. Visualization done with [28].

in the range [50, 500]. The EDs are located according to a

random and uniform distribution in a target area. In this work,
we assume the target area coincides with the Luxembourg
country (Fig.6). We derived the exact location of the EDs
using random uniform distribution in the GEOPY [29] library
of Python programming language. Besides the considered
scenario, the algorithm is not limited to this selected geo-
graphical area and applies to any other one.

FIGURE 9. Packet collisions during the satellite visibility in one month

with 1 and 2 satellites.

analysis, we adopted their real visibility over Luxembourg

during October 2021. Moreover, we assumed an elevation
angle 2e = 30◦ for both satellites.
We used SKYFIELD library [30] with the CelesTrak [31]
online Two Line Element (TLE) sources for generating the
real visibility time tables of the satellites per each ED. A TLE
provides the satellite location along its orbit for a given point-
ing time (the time epoch), by encoding the orbital elements.
The specific location of the EDs (uniformly distributed over
the Country of Luxembourg), the satellites TLE, and the
fixed elevation angle (2e = 30◦ ) were given in input to
the SKYFIELD. Finally, the obtained human-readable time
FIGURE 7. Orbits of LacunaSat 3 and LacunaSat 2B satellites compared to outputs were translated into epoch timestamps and used in the
the target area on the Earth, Luxembourg. Matlab algorithm, implementing SALSA. The code is public
available at [32].
We considered the Lacuna satellites, in particular Lacuna- We assume the Guard Time Tg is equal to 10ms; and the
Sat 3 and Lacuna-Sat 2B which have the approximate height ToA is the maximum allowed ToA in the EU868 region, with
of 500Km to 600Km from the earth. In the performance SF = 12 and the maximum payload of 51 bytes. It follows,

VOLUME 10, 2022 11613

 M. Afhamisis, M. R. Palattella: SALSA: Scheduling Algorithm for LoRa to LEO Satellites

knowledge about the satellite visibility. Note that we

introduced a random uniformly distributed offset in the
start time of the first uplink per each ED.
• SALSA w/ FCFS: the EDs that are visited first by the
satellites transmit first, but only when the channel is not
busy (no collision due to overlapping transmissions).
• SALSA w/ FAIR: all the EDs get an equal chance to
transmit thanks to the FAIR policy, in addition to FCFS.
When (i) the EDs are not aware of the visibility time of the
satellite(s), and (ii) no scheduling algorithm is implemented,
the network faces a high number of packet drops and packet
collisions. Packet drops are due to the lack of availability
of the satellite gateway, while packet collisions are due to
concurrent random transmissions. As shown in Fig. 8, the
adoption of a constellation of two satellites does not improve
the network performance in the absence of a scheduling
technique. As depicted in Fig. 9, the performance in term of
the number of collisions get even worst with two satellites: it
is almost double with a large network size.
Contrary to random ALOHA-like access, when adopting
SALSA the LNS schedules the transmissions while aware of
the location of the EDs and the satellite availability time for
each of them.
We compared the performance achievable with FCFS and
FAIR policy in the worst-case scenario, with a single LEO
satellite, and with a constellation of two satellites (Fig. 10).
FCFS policy schedules the transmissions according only to
location-based priority. It follows that some devices (those
visited last by the satellite) never get a chance of an uplink
opportunity. Moreover, each device may get a different num-
ber of reserved slots (as confirmed by the large whiskers and
the many outliers). This behaviour is more pronounced with
two satellites. On the contrary, the FAIR policy guarantees
all devices get the same chance, even in large networks,
besides the size of the constellation. With two satellites, the
network performance improves, and the average number of
uplink transmissions per ED is almost duplicated compared
to the scenario with a single LEO satellite. The obtained
results confirm the reliability and scalability of the SALSA
algorithm with the FAIR policy.

V. CONCLUSION
In the present work, we tackle the issue of a LoRaWAN
FIGURE 10. Average Number of Uplink Transmission for each ED, with mobile gateway installed on the LEO satellite. For the first
FCFS and FAIR policy (a) with one satellite, and (b) with two satellites.
time, we propose a TDMA scheduling technique, SALSA.
It optimises the usage of the satellite bandwidth during
limited satellite visibility. The proposed SALSA algorithm
the time reserved for each ED, for a single transmission is first avoids by design any collision, thanks to the TDMA
Tr = 2 × Tg + ToA = 2813.5 ms. approach; second, it avoids packet drop by considering the
satellite availability; third, it ensures fair access for all
B. ACHIEVABLE PERFORMANCE the end devices in the network. With simulation results,
We have compared the behaviour of the satellite LoRaWAN we demonstrate in a realistic scenario that the SALSA algo-
network in three different scenarios: rithm can improve the network performance. The opportu-
• ALOHA: the EDs transmit following the LoRa stan- nities of uplink transmission per each ED decreases with
dard, with random ALOHA access. They transmit increasing network size (more EDs are contending for the
periodically, every 30 minutes, without having any same channel). LEO constellations help overcome this issue.

11614 VOLUME 10, 2022

M. Afhamisis, M. R. Palattella: SALSA: Scheduling Algorithm for LoRa to LEO Satellites

The transmission opportunities are almost duplicated when [21] D. Zorbas, ‘‘A testbed for time-slotted LoRa communications,’’ in Proc.
considering two satellites, and they could be further improved 21st Int. Symp. World Wireless, Mobile Multimedia Netw., Aug. 2020,
pp. 182–184.
with larger constellations. In our future work, we will extend [22] D. Zorbas, C. Caillouet, K. A. Hassan, and D. Pesch, ‘‘Optimal data
the SALSA algorithm to support both uplink and downlink collection time in LoRa networks—A time-slotted approach,’’ Sensors,
traffic, with Class A and B devices. vol. 21, no. 4, p. 1193, Feb. 2021.
[23] M. Asad Ullah, K. Mikhaylov, and H. Alves, ‘‘Enabling mMTC in remote
areas: LoRaWAN and LEO satellite integration for offshore wind farms
REFERENCES monitoring,’’ IEEE Trans. Ind. Informat., early access, Sep. 14, 2021, doi:
[1] I. F. Akyildiz and A. Kak, ‘‘The internet of space things/cubesats,’’ IEEE 10.1109/TII.2021.3112386.
Netw., vol. 33, no. 5, pp. 212–218, Sep. 2019. [24] F. Tondo, S. Montejo-Sánchez, M. Pellenz, S. Céspedes, and R. Souza,
[2] LoRa Alliance. LoRaWAN Specifications. Accessed: Sep. 1, 2021. ‘‘Direct-to-satellite IoT slotted aloha systems with multiple satellites and
[Online]. Available: https://lora-alliance.org/ unequal erasure probabilities,’’ Sensors, vol. 21, no. 21, p. 7099, 2021.
[3] Lacuna Space Website. Accessed: Mar. 22, 2021. [Online]. Available: [25] SEMTECH. (Feb. 2020). LoRa and LoRaWAN: A Technical Overview.
https://lacuna.space/ [Online]. Available: https://www.semtech.com/lora
[4] Wyld Networks. Creating the Networks That Matter. [26] T. Polonelli, D. Brunelli, A. Marzocchi, and L. Benini, ‘‘Slotted ALOHA
Accessed: Mar. 16, 2021. [Online]. Available: https://https:// on LoRaWAN-design, analysis, and deployment,’’ Sensors, vol. 19, no. 4,
wyldnetworks.com/wyld-announces-global-lora-satellite-deal/ p. 838, Feb. 2019.
[5] M. Centenaro, C. E. Costa, F. Granelli, C. Sacchi, and L. Vangelista, [27] The Things Network. ToA Calculator. [Online]. Available: https://www.
‘‘A survey on technologies, standards and open challenges in satellite thethingsnetwork.org/airtime-calculator
IoT,’’ IEEE Commun. Surveys Tuts., vol. 23, no. 3, pp. 1693–1720, [28] MapBox Studio. Accessed: Nov. 31, 2021. [Online]. Available:
3rd Quart., 2021. https://www.mapbox.com/
[6] M. R. Palattella and N. Accettura, ‘‘Enabling internet of everything every- [29] GEOPY, a Python Client for Several Popular Geocoding Web Ser-
where: LPWAN with satellite backhaul,’’ in Proc. Global Inf. Infrastruct. vices. Accessed: Nov. 31, 2021. [Online]. Available: https://github.
Netw. Symp., Oct. 2018, pp. 1–5. com/geopy/geopy
[7] L. Chaari, M. Fourati, and J. Rezgu, ‘‘Heterogeneous LoRaWAN LEO [30] Skyfield Repository. Accessed: Nov. 1, 2021. [Online]. Available:
satellites networks concepts, architectures and future directions,’’ in Proc. https://github.com/skyfielders/python-skyfield
Global Inf. Infr. Net. Symp. (GIIS), 2019, pp. 1–6. [31] CelesTrak. Accessed: Nov. 1, 2021. [Online]. Available: https://celestrak.
[8] Z. Qu, G. Zhang, H. Cao, and J. Xie, ‘‘LEO satellite constellation for com/
Internet of Things,’’ IEEE Access, vol. 5, pp. 18391–18401, 2017. [32] SALSA Algorithm. Accessed: Jan. 22, 2022. [Online]. Available:
[9] G. Boquet, P. Tuset-Peiro, F. Adelantado, T. Watteyne, and X. Vilajosana, https://github.com/list-luxembourg/salsa
‘‘LR-FHSS: Overview and performance analysis,’’ IEEE Commun. Mag.,
vol. 59, no. 3, pp. 30–36, Mar. 2021.
[10] A. Doroshkin, A. Zadorozhny, O. Kus, V. Prokopyev, and Y. Prokopyev, MOHAMMAD AFHAMISIS (Member, IEEE)
‘‘Experimental study of LoRa modulation immunity to Doppler effect in received the bachelor’s degree in telecommuni-
CubeSat radio communications,’’ IEEE Access, vol. 7, pp. 75721–75731, cation engineering from Semnan University, Iran,
2019. and the Master of Science degree in satellite tech-
[11] G. Ferre, ‘‘Collision and packet loss analysis in a LoRaWAN network,’’ in nology engineering from the Iran University of
Proc. Eur. Signal Process. Conf. (EUSIPCO), Aug. 2017, pp. 2589–2590. Science and Technology (IUST). He is currently
[12] P. Gkotsiopoulos, D. Zorbas, and C. Douligeris, ‘‘Performance determi- pursuing the Ph.D. degree with the ERIN Depart-
nants in LoRa networks: A literature review,’’ IEEE Commun. Surveys
ment, Luxembourg Institute of Science and Tech-
Tuts., vol. 23, no. 3, pp. 1721–1758, 3rd Quart., 2021.
nology (LIST), working on optimizing LoRaWAN
[13] N. Chinchilla-Romero, J. Navarro-Ortiz, P. Mu noz, and P. Ameigeiras,
‘‘Collision avoidance resource allocation for LoRaWAN,’’ Sensors, protocol for integration with satellite for smart
vol. 21, no. 4, p. 1218, Feb. 2021. agriculture applications. His thesis was on optimizing the control of GEO
[14] M. Alenezi, K. Chai, A. Alam, Y. Chen, and S. Jimaa, ‘‘Unsupervised satellites with large antennas and deployable solar cells. Also, he has six
learning clustering and dynamic transmission scheduling for efficient years of experience as a Senior Radio Communications Specialist and man-
dense LoRaWAN networks,’’ IEEE Access, vol. 8, pp. 191495–191509, aging different projects on radio communications and the IoT monitoring
2020. topics in the Iranian Railways Headquarter Office.
[15] G. Yapar, T. Tugcu, and O. Ermis, ‘‘Time-slotted ALOHA-based
LoRaWAN scheduling with aggregated acknowledgement approach,’’ in
Proc. 25th Conf. Open Innov. Assoc., Nov. 2019, pp. 383–390. MARIA RITA PALATTELLA (Member, IEEE)
[16] G. Lee and J. Youn, ‘‘Group-based transmission scheduling scheme for received the Ph.D. degree, in 2011. She is currently
building LoRa-based massive IoT,’’ in Proc. Int. Conf. Artif. Intell. Inf. a Senior Research and Technology Associate at
Commun., Feb. 2020, pp. 583–586. the LIST. She is leading the work on designing
[17] J. Pullmann and D. Macko, ‘‘A new planning-based collision-prevention innovative network architectures and cross-layer
mechanism in long-range IoT networks,’’ IEEE Internet Things J., vol. 6, protocols optimization for the Internet of Things
no. 6, pp. 9439–9446, Dec. 2019. applications for environmental monitoring, preci-
[18] C. E. Fehri, M. Kassab, N. Baccour, S. Abdellatif, P. Berthou, and
sion agriculture, and disaster management. She
I. Kammoun, ‘‘An uplink synchronization scheme for LoRaWAN class B,’’
is investigating means of integrating satellite and
in Proc. Int. Conf. Wireless Mobile Comput., Netw. Commun., Oct. 2019,
pp. 47–52. terrestrial networks for future high-reliable end-
[19] L. Chasserat, N. Accettura, and P. Berthou, ‘‘Short: Achieving energy to-end communication systems. She is a member of IEEE ComSoc WISE,
efficiency in dense LoRaWANs through TDMA,’’ in Proc. 21st Int. Symp. and IETF 6TiSCH WG, and the Secretary of the IEEE 5G Mobile Wireless
World Wireless, Mobile Multimedia Netw., Aug. 2020, pp. 26–29. Internet Technical Subcommittee. She sits on the Editorial Board of the
[20] K. Q. Abdelfadeel, D. Zorbas, V. Cionca, and D. Pesch, ‘‘FREE—Fine- Transactions on Emerging Telecommunications Technologies and the EAI
grained scheduling for reliable and energy-efficient data collection in Transactions on IoT.
LoRaWAN,’’ IEEE Internet Things J., vol. 7, no. 1, pp. 669–683, Jan. 2020.

VOLUME 10, 2022 11615