Content uploaded by Kais Mekki
Author content
All content in this area was uploaded by Kais Mekki on Feb 08, 2019
Content may be subject to copyright.
Please cite this article in press as: K. Mekki, et al., A comparative study of LPWAN technologies for large-scale IoT deployment, ICT Express (2018), https://doi.org/10.1016/j.icte.2017.12.005.
Available online at www.sciencedirect.com
ScienceDirect
ICT Express ( ) –
www.elsevier.com/locate/icte
A comparative study of LPWAN technologies for large-scale IoT
deployment
Kais Mekkia,∗, Eddy Bajica, Frederic Chaxela, Fernand Meyerb
aResearch Centre for Automatic Control of Nancy, Campus Sciences, BP 70239, Vandoeuvre, 54506, France
bOKKO SAS, 34 Rue Nationale, Puttelange-aux-Lacs, 57510, France
Received 21 September 2017; accepted 20 December 2017
Available online xxxx
Abstract
By 2020, more than 50 billion devices will be connected through radio communications. In conjunction with the rapid growth of the Internet
of Things (IoT) market, low power wide area networks (LPWAN) have become a popular low-rate long-range radio communication technology.
Sigfox, LoRa, and NB-IoT are the three leading LPWAN technologies that compete for large-scale IoT deployment. This paper provides a
comprehensive and comparative study of these technologies, which serve as efficient solutions to connect smart, autonomous, and heterogeneous
devices. We show that Sigfox and LoRa are advantageous in terms of battery lifetime, capacity, and cost. Meanwhile, NB-IoT offers benefits in
terms of latency and quality of service. In addition, we analyze the IoT success factors of these LPWAN technologies, and we consider application
scenarios and explain which technology is the best fit for each of these scenarios.
c
⃝2018 The Korean Institute of Communications Information Sciences. Publishing Services by Elsevier B.V. This is an open access article under
the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Internet of Things; LPWAN; LoRa; Sigfox; NB-IoT
1. Introduction
The Internet of Things (IoT) refers to the inter connection
and exchange of data among devices/sensors. Currently, with
the explosive growth of the IoT technologies, an increasing
number of practical applications can be found in many fields
including security, asset tracking, agriculture, smart metering,
smart cities, and smart homes [1]. IoT applications have specific
requirements such as long range, low data rate, low energy con-
sumption, and cost effectiveness. The widely used short-range
radio technologies (e.g., ZigBee, Bluetooth) are not adapted
for scenarios that require long range transmission. Solutions
based on cellular communications (e.g., 2G, 3G, and 4G) can
provide larger coverage, but they consume excessive device
energy. Therefore, IoT applications’ requirements have driven
the emergence of a new wireless communication technology:
low power wide area network (LPWAN).
∗Corresponding author.
E-mail address: kais.mekki@univ-lorraine.fr (K. Mekki).
Peer review under responsibility of The Korean Institute of Communica-
tions Information Sciences.
LPWAN is increasingly gaining popularity in industrial and
research communities because of its low power, long range,
and low-cost communication characteristics. It provides long-
range communication up to 10–40 km in rural zones and 1–5
km in urban zones [2]. In addition, it is highly energy efficient
(i.e. 10+ years of battery lifetime [3]) and inexpensive, with
the cost of a radio chipset being less than 2eand an operating
cost of 1eper device per year [4]. These promising aspects of
LPWAN have prompted recent experimental studies on the per-
formance of LPWAN in outdoor and indoor environments [5–
7]. In summary, LPWAN is highly suitable for IoT applications
that only need to transmit tiny amounts of data in long range, as
shown in Fig. 1. As recently as early 2013, the term “LPWAN”
did not even exist [8]. Many LPWAN technologies have arisen
in the licensed as well as unlicensed frequency bandwidth.
Among them, Sigfox, LoRa, and NB-IoT are today’s leading
emergent technologies that involve many technical differences.
The Sigfox technology was developed in 2010 by the start-
up Sigfox (in Toulouse, France), which is both a company
https://doi.org/10.1016/j.icte.2017.12.005
2405-9595/ c
⃝2018 The Korean Institute of Communications Information Sciences. Publishing Services by Elsevier B.V. This is an open access article under the
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article in press as: K. Mekki, et al., A comparative study of LPWAN technologies for large-scale IoT deployment, ICT Express (2018), https://doi.org/10.1016/j.icte.2017.12.005.
2K. Mekki et al. / ICT Express ( ) –
Fig. 1. Required data rate vs. range capacity of radio communication technolo-
gies: LPWAN positioning.
and an LPWAN network operator. Sigfox operates and com-
mercializes its own IoT solution in 31 countries and is still
under rollout worldwide owing to the partnership with various
network operators [9]. LoRa was first developed by the start-up
Cycleo in 2009 (in Grenoble, France) and was purchased three
years later by Semtech (USA). In 2015, LoRa was standardized
by LoRa-Alliance and is deployed in 42 countries and is still
under rollout in other countries owing to the investment of
various mobile operators (e.g., Bouygues and Orange in France,
KPN in Netherlands, and Fastnet in South Africa) [10].
NB-IoT is an LPWAN technology based on narrow band
radio technology and is standardized by the 3rd generation
partnership project (3GPP). Its specifications were published
in Release 13 of the 3GPP on June 2016. The NB-IoT is still
under test in Europe. In December 2016, Vodafone and Huawei
integrated NB-IoT into the Spanish Vodafone network and sent
the first message conforming to the NB-IoT standard to a device
installed in a water meter. Currently, Huawei is multiplying
partnerships to deploy this technology worldwide (it was an-
nounced to be deployed in many countries in 2018). In May
2017, the Ministry of Industry and Information Technology in
China announced its decision to accelerate the commercial use
of NB-IoT for utilities and smart city applications.
In this paper, the technical differences of Sigfox, LoRa,
and NB-IoT are presented and compared in terms of physi-
cal/communication features. In addition, these technologies are
compared in terms of IoT success factors such as quality of
service (QoS), coverage, range, latency, battery life, scalability,
payload length, deployment, and cost. Further, we consider
application scenarios and explain which technology fits best.
The remainder of this paper is organized as follows: Sec-
tion 2describes the technical features of Sigfox, LoRa, and NB-
IoT. Section 3compares them in terms of IoT factors. Section 4
explains which technology fits best for different application
scenarios. Finally, Section 5discusses and concludes the paper.
2. Technical differences: SIGFOX, LORA, and NB-IOT
In this section, we highlight the emerging proprietary tech-
nologies and the technical aspects of Sigfox, LoRa, and NB-IoT
as summarized in Table 1.
2.1. Sigfox
Sigfox is an LPWAN network operator that offers an end-
to-end IoT connectivity solution based on its patented tech-
nologies. Sigfox deploys its proprietary base stations equipped
with cognitive software-defined radios and connect them to the
back end servers using an IP-based network. The end devices
connected to these base stations using binary phase-shift keying
(BPSK) modulation in an ultra-narrow band (100 Hz) sub-
GHZ ISM band carrier. Sigfox uses unlicensed ISM bands, for
example, 868 MHz in Europe, 915 MHz in North America, and
433 MHz in Asia. By employing the ultra-narrow band, Sigfox
uses the frequency bandwidth efficiently and experiences very
low noise levels, leading to very low power consumption, high
receiver sensitivity, and low-cost antenna design at the expense
of maximum throughput of only 100 bps. Sigfox initially
supported only uplink communication, but later evolved to
bidirectional technology with a significant link asymmetry. The
downlink communication, i.e., data from the base stations to the
end devices can only occur following an uplink communication.
The number of messages over the uplink is limited to 140 mes-
sages per day. The maximum payload length for each uplink
message is 12 bytes. However, the number of messages over the
downlink is limited to four messages per day, which means that
the acknowledgment of every uplink message is not supported.
The maximum payload length for each downlink message is
eight bytes. Without the adequate support of acknowledgments,
the uplink communication reliability is ensured using time and
frequency diversity as well as transmission duplication. Each
end-device message is transmitted multiple times (three by
default) over different frequency channels. For this purpose,
in Europe for example, the band between 868.180 MHz and
868.220 MHz is divided into 400 orthogonal 100 Hz channels
(among them 40 channels are reserved and not used) [4]. As
the base stations can receive messages simultaneously over all
channels, the end device can randomly choose a frequency
channel to transmit their messages. This simplifies the end-
device design and reduces its cost.
2.2. LoRa
LoRa is a physical layer technology that modulates the sig-
nals in sub-GHZ ISM band using a proprietary spread spectrum
technique [11]. Like Sigfox, LoRa uses unlicensed ISM bands,
i.e., 868 MHz in Europe, 915 MHz in North America, and
433 MHz in Asia. The bidirectional communication is provided
by the chirp spread spectrum (CSS) modulation that spreads
a narrow-band signal over a wider channel bandwidth. The
resulting signal has low noise levels, enabling high interference
resilience, and is difficult to detect or jam [12].
LoRa uses six spreading factors (SF7 to SF12) to adapt the
data rate and range tradeoff. Higher spreading factor allows
longer range at the expense of lower data rate, and vice versa.
The LoRa data rate is between 300 bps and 50 kbps depending
on spreading factor and channel bandwidth. Further, messages
transmitted using different spreading factors can be received
simultaneously by LoRa base stations [13]. The maximum
Please cite this article in press as: K. Mekki, et al., A comparative study of LPWAN technologies for large-scale IoT deployment, ICT Express (2018), https://doi.org/10.1016/j.icte.2017.12.005.
K. Mekki et al. / ICT Express ( ) – 3
Table 1
Overview of LPWAN technologies: Sigfox, LoRa, and NB-IoT.
Sigfox LoRaWAN NB-IoT
Modulation BPSK CSS QPSK
Frequency Unlicensed ISM bands (868 MHz in Europe, 915
MHz in North America, and 433 MHz in Asia)
Unlicensed ISM bands (868 MHz in Europe, 915
MHz in North America, and 433 MHz in Asia)
Licensed LTE frequency
bands
Bandwidth 100 Hz 250 kHz and 125 kHz 200 kHz
Maximum data rate 100 bps 50 kbps 200 kbps
Bidirectional Limited / Half-duplex Yes / Half-duplex Yes / Half-duplex
Maximum messages/day 140 (UL), 4 (DL) Unlimited Unlimited
Maximum payload length 12 bytes (UL), 8 bytes (DL) 243 bytes 1600 bytes
Range 10 km (urban), 40 km (rural) 5 km (urban), 20 km (rural) 1 km (urban), 10 km
(rural)
Interference immunity Very high Very high Low
Authentication & encryption Not supported Yes (AES 128b) Yes (LTE encryption)
Adaptive data rate No Yes No
Handover End-devices do not join a single base station End-devices do not join a single base station End-devices join a
single base station
Localization Yes (RSSI) Yes (TDOA) No (under specification)
Allow private network No Yes No
Standardization Sigfox company is collaborating with ETSI on
the standardization of Sigfox-based network
LoRa-Alliance 3GPP
Fig. 2. Bidirectional communication between end-device and base station for LoRaWAN class A.
payload length for each message is 243 bytes. A LoRa-based
communication protocol called LoRaWAN was standardized by
LoRa-Alliance (first version in 2015). Using LoRaWAN, each
message transmitted by an end device is received by all the base
stations in the range. By exploiting this redundant reception,
LoRaWAN improves the successfully received messages ratio.
However, achieving this feature requires multiple base stations
in the neighborhood, which may increase the network deploy-
ment cost. The resulting duplicate receptions are filtered in
the backend system (network server) that also has the required
intelligence for checking security, sending acknowledgments to
the end device, and sending the message to the corresponding
application server. Further, multiple receptions of the same
message by different base stations are exploited by LoRaWAN
for localizing end devices. For this purpose, the time difference
of arrival (TDOA)-based localization technique supported by
very accurate time synchronization between multiple base sta-
tions is used. Moreover, multiple receptions of the same mes-
sage at different base stations avoid the handover in LoRaWAN
network (i.e., if a node is mobile or moving, handover is not
needed between the base stations).
In addition, LoRaWAN provides various classes of end
devices to address the different requirements of a wide range
of IoT applications, e.g., latency requirements.
–Bidirectional end devices (class A): class-A end devices
allow bidirectional communications where by each end-
device’s uplink transmission is followed by two short
downlink receive windows as shown in Fig. 2. The
transmissions lot scheduled by the end device is based
on its own communication needs with a small variation
based on a random time basis. This class-A operation
is the lowest power end-device system for applications
that only require short downlink communication after
the end device has sent an uplink message. Downlink
communications at any other time will have to wait until
the next uplink message of the end device.
–Bidirectional end devices with scheduled receives lots
(class B): in addition to the random receive windows
of class A, class B devices open extra receive windows
at scheduled times. To open receive windows at the
scheduled time, end devices receive a time-synchronized
beacon from the base station. This allows the network
server to know when the end device is listening.
–Bidirectional end devices with maximal receive slots
(class C): class C end devices have almost continuously
open receive windows, and only close when transmitting
at the expense of excessive energy consumption.
The specifications of the next version of LoRaWAN are
still being developed by LoRa-Alliance [10]. The new features
expected are roaming, class-B clarification, and the temporary
switching between class A and class C.
2.3. NB-IoT
NB-IoT is a Narrow Band IoT technology specified in
Release 13 of the 3GPP in June 2016. NB-IoT can coexist
with GSM (global system for mobile communications) and LTE
(long-term evolution) under licensed frequency bands (e.g., 700
Please cite this article in press as: K. Mekki, et al., A comparative study of LPWAN technologies for large-scale IoT deployment, ICT Express (2018), https://doi.org/10.1016/j.icte.2017.12.005.
4K. Mekki et al. / ICT Express ( ) –
Fig. 3. Operation modes for NB-IoT.
MHz, 800 MHz, and 900 MHz). NB-IoT occupies a frequency
band width of 200 KHz, which corresponds to one resource
block in GSM and LTE transmission [14]. With this frequency
band selection, the following operation modes are possible, as
shown in Fig. 3:
–Stand-alone operation: a possible scenario is the utiliza-
tion of GSM frequencies bands currently used.
–Guard-band operation: utilizing the unused resource
blocks within an LTE carrier’s guard band.
–In-band operation: utilizing resource blocks within an
LTE carrier.
For the stand-alone operation, the GSM carriers in the right
part of Fig. 3 are shown as an example to indicate that the
operation is possible in NB-IoT deployment. In fact, the 3GPP
recommends the integration of NB-IoT in conjunction with the
LTE cellular networks. NB-IoT can be supported with only a
software upgrade in addition to the existing LTE infrastructure.
The NB-IoT communication protocol is based on the LTE
protocol. In fact, NB-IoT reduces LTE protocol functionalities
to the minimum and enhances them as required for IoT ap-
plications. For example, the LTE backend system is used to
broadcast information that is valid for all end devices within
a cell. As the broadcasting back end system obtains resources
and consumes battery power from each end device, it is kept
to a minimum, in size as well as in its occurrence. It was
optimized to small and infrequent data messages and avoids the
features not required for the IoT purpose, e.g., measurements
to monitor the channel quality, carrier aggregation, and dual
connectivity. Therefore, the end devices require only a small
amount of battery, thus making it cost-efficient.
Consequently, NB-IoT technology can be regarded as a
new air interface from the protocol stack point of view, while
being built on the well-established LTE infrastructure. NB-IoT
allows connectivity of up to 100 K end devices per cell with
the potential for scaling up the capacity by adding more NB-
IoT carriers. NB-IoT uses the single-carrier frequency division
multiple access (FDMA) in the uplink and orthogonal FDMA
(OFDMA) in the downlink, and employs the quadrature phase-
shift keying modulation (QPSK) [14]. The data rate is limited
to 200 kbps for the downlink and to 20 kbps for the uplink.
The maximum payload size for each message is 1600 bytes.
As discussed in [15], NB-IoT technology can achieve 10 years
of battery lifetime when transmitting 200 bytes per day on
average.
The improvement of NB-IoT continues with Release 15
of the 3GPP. According to the 3GPP’s current plan, the NB-
IoT will be extended to include localization methods, multi-
cast services (e.g., end-devices software update and messages
concerning a whole group of end devices), mobility, as well
as further technical details to enhance the applications of the
NB-IoT technology.
3. Comparison in terms of IoT factors
Many factors should be considered when choosing the ap-
propriate LPWAN technology for an IoT application including
quality of service, battery life, latency, scalability, payload
length, coverage, range, deployment, and cost. In the following,
Sigfox, LoRa and NB-IoT are compared in terms of these
factors and their technical differences.
3.1. Quality of service
Sigfox and LoRa employ unlicensed spectra and asyn-
chronous communication protocols. They can bounce interfer-
ence, multipath, and fading. However, they cannot offer the
same QoS provided by NB-IoT. NB-IoT employs a licensed
spectrum and an LTE-based synchronous protocol, which are
optimal for QoS at the expense of cost, i.e., licensed LTE
spectrum auctions are over 500 million euro per MHz [8].
Owing to QoS and cost tradeoff, NB-IoT is preferred for
applications that require guaranteed quality of service, whereas
applications that do not have this constraint should choose
LoRa or Sigfox.
3.2. Battery life & Latency
In Sigfox, LoRa, and NB-IoT, end devices are in sleep mode
most of the time outside operation, which reduce the amount
of consumed energy, i.e., long end-devices lifetime. However,
the NB-IoT end device consumes additional energy because
of synchronous communication and QoS handling, and its
OFDM/FDMA access modes require more peak current [16].
This additional energy consumption reduces the NB-IoT end-
device lifetime as compared to Sigfox and LoRa.
However, NB-IoT offers the advantage of low latency.
Unlike Sigfox, LoRa provides class C to also handle low-
bidirectional latency at the expense of increased energy con-
sumption. Therefore, for applications that are insensitive to the
latency and do not have large amount of data to send, Sigfox and
class-A LoRa are the best options. For applications that require
low latency, NB-IoT and class-C LoRa are the better choices.
3.3. Scalability & Payload length
The support of the massive number of devices is one of the
key features of Sigfox, LoRa, and NB-IoT. These technologies
Please cite this article in press as: K. Mekki, et al., A comparative study of LPWAN technologies for large-scale IoT deployment, ICT Express (2018), https://doi.org/10.1016/j.icte.2017.12.005.
K. Mekki et al. / ICT Express ( ) – 5
Table 2
Different costs of Sigfox, LoRa, and NB-IoT.
Spectrum cost Deployment cost End-device cost
Sigfox Free >4000e/base station <2e
LoRa Free >100e/gateway >1000e/base station 3–5e
NB-IoT >500 Me/MHz >15 000e/base station >20e
work well with the increasing number and density of connected
devices. Several techniques are considered to cope with this
scalability feature such as the efficient exploitation of diversity
in a channel, as well as in time and space. However, NB-IoT
offers the advantage of very high scalability than Sigfox and
LoRa. NB-IoT allows connectivity of up to 100 K end devices
per cell compared to 50 K per cell for Sigfox and LoRa [13].
Nevertheless, NB-IoT also offers the advantage of maximum
payload length. As presented in Table 1, NB-IoT allows the
transmission of data of up to 1600 bytes. LoRa allows a
maximum of 243 bytes of data to be sent. In contrary, Sigfox
proposes the lowest payload length of 12 bytes, which limits its
utilization on various IoT applications that need to send large
data sizes.
3.4. Network coverage & Range
The major utilization advantage of Sigfox is that an entire
city can be covered by one single base station (i.e., range
>40 km). In Belgium, a country with a total surface area of
approximately 30 500 km2, the Sigfox network deployment
covers the entire country with only seven base stations [8].
By contrast, LoRa has a lower range (i.e., range <20 km)
that requires only three base stations to cover an entire city
such as Barcelona. NB-IoT has the lowest range and coverage
capabilities (i.e., range <10 km). It focuses primarily on the
class of devices that are installed at places far from the typical
reach of cellular networks (e.g., indoors, deep indoors). In
addition, the deployment of NB-IoT is limited to LTE base
stations. Thus, it is not suitable for rural or suburban regions
that do not benefit from LTE coverage.
3.5. Deployment model
The NB-IoT specifications were released in June 2016; thus,
additional time will be needed before its network is established.
However, the Sigfox and LoRa ecosystems are mature and
are now under commercialization in various countries and
cities. LoRa has the advantage that allows it to be currently
deployed in 42 countries versus 31 countries for Sigfox [9,10].
Nevertheless, the world wide deployments of LoRa and Sigfox
are still under rollout.
In addition, one significant advantage of LoRa ecosystem
is its flexibility. Unlike Sigfox and NB-IoT, LoRa offers local
network deployment, i.e., LAN using LoRa gateway as well
as public network operation via base stations. In the industrial
field, a hybrid operating model could be used to deploy a
local LoRa network in factory areas and uses the public LoRa
network to cover the outside areas.
Fig. 4. Respective advantages of Sigfox, LoRa, and NB-IoT in terms of IoT
factors.
3.6. Cost
Various cost aspects need to be considered such as spectrum
cost (license), network/deployment cost, and device cost. Ta-
ble 2 shows the cost of Sigfox, LoRa, and NB-IoT. It is apparent
that Sigfox and LoRa are more cost-effective compared to NB-
IoT.
In summary, Sigfox, LoRa, and NB-IoT each has their
respective advantages in terms of different IoT factors as shown
in Fig. 4.
4. Application examples: which technology fits best?
The IoT factors and technical differences of Sigfox, LoRa,
and NB-IoT will determine their feasibility for specific applica-
tions. As discussed in this paper, one technology cannot equally
serve all IoT applications. In this section, various application
use cases are discussed with a summary of the best-fitting
technology.
4.1. Electric metering
In the electric metering market, companies typically require
frequent communication, low latency, and high data rate. Typ-
ically, they require neither low energy consumption nor long
battery life time as electric meters have continuous power
source. Moreover, companies need real-time grid monitoring
to make immediate decisions, e.g., load, outages, and interrup-
tions. Thus, Sigfox is inappropriate for this application as it
does not handle low latency. On the contrary, electric meters
can be setup using class-C LoRa to ensure very low latency.
However, NB-IoT is a better fit for this application because
of the required high data rate and frequent communication.
Moreover, electric meters are typically found in stationary
locations in densely populated areas. Therefore, it is easy to
ensure NB-IoT coverage by cellular operators (LTE).
Please cite this article in press as: K. Mekki, et al., A comparative study of LPWAN technologies for large-scale IoT deployment, ICT Express (2018), https://doi.org/10.1016/j.icte.2017.12.005.
6K. Mekki et al. / ICT Express ( ) –
4.2. Smart farming
In agriculture, the long battery lifetime of sensor devices
are required. Temperature, humidity, and alkalinity sensors
could significantly reduce water consumption and improve
yield. The devices update sensed data a few times per hour as
the environment conditions have not radically changed. Thus,
Sigfox and LoRa are ideal for this application. Moreover, many
farms today do not have LTE cellular coverage; thus, NB-IoT is
not the solution for agriculture in the near future.
4.3. Manufacturing automation
Real-time machinery monitoring prevents industrial pro-
duction line down and allows remote control to improve ef-
ficiency. In factory automation, various types of sensors and
communication requirements exist. Some applications require
frequent communication and high-quality service, thus NB-IoT
is a better solution than Sigfox and LoRa. Other applications
require low-cost sensors and long battery lifetime for asset
tracking and status monitoring; in this case, Sigfox and LoRa
are a better solution. Because of the various requirements,
hybrid solutions could also be used.
4.4. Smart building
Temperature, humidity, security, water flow, and electric
plugs sensors alert property managers to prevent damages and
instantly respond to requests without having a manual building
monitor. The buildings’ cleaning and usage could also be
carried out more efficiently. These sensors require low cost and
long battery lifetime. They do not require quality of service
or frequent communication, therefore Sigfox and LoRa are a
better fit for this class of applications.
4.5. Retail point of sale terminals
Sale-point systems require guaranteed quality of service
as they handle frequent communications. These systems have
continuous electrical power source, thus there is no constraint
on battery lifetime. There is also a strong requirement of low
latency, i.e., long latency times limit the number of transactions
that a store can make. Thus, NB-IoT is a better fit for this
application.
4.6. Pallet tracking for logistics
Currently, pallets tracking to determine the goods’ location
and condition are highly desirable in logistics. In this appli-
cation, the most sought-after requirements are device cost and
battery lifetime. Pallet tracking is a good example of a hybrid-
deployment solution. Logistics companies can have their own
network to ensure guaranteed coverage in their facilities. Low-
cost IoT devices could be easily deployed on vehicles. Sigfox
or LoRa public base stations can then be used when vehicles are
outside the facilities or when goods arrive at customer locations.
However, LoRa allows more reliable communications than
Sigfox when moving at high speeds [3]. For NB-IoT, the LTE
network might not be available in all logistic locations, typically
in rural areas. Owing to the low cost, long battery lifetime, and
reliable mobile communications, LoRa is a better fit for this
application.
5. Conclusion
This paper has summarized the technical differences of
Sigfox, LoRa, and NB-IoT, and discussed their advantages in
terms of IoT factors and major issues. Each technology will
have its place in the IoT market. Sigfox and LoRa will serve
as the lower-cost device, with very long range (high coverage),
infrequent communication rate, and very long battery lifetime.
Unlike Sigfox, LoRa will also serve the local network deploy-
ment and the reliable communication when devices move at
high speeds. By contrast, NB-IoT will serve the higher-value
IoT markets that are willing to pay for very low latency and
high quality of service.
Despite the cellular companies’ tests, the lack of NB-IoT
commercial deployments currently leaves open questions on
the actual battery lifetime and the performance attainable by
this technology in real-world conditions. Finally, it is expected
that 5th generation (5G) wireless mobile communication will
provide the means to allow an all-connected world of humans
and devices by the year 2020, which would lead to a global
LPWAN solution for IoT applications.
Conflict of interest
All the authors declare that there is no conflict of interest
regarding the publication of this paper.
References
[1] R. Ratasuk, N. Mangalvedhe, A. Ghosh, Overview of LTE enhancements
for cellular IoT, in: Proc. of PIMRC, Hong Kong, China, 2015, pp. 2293–
2297.
[2] M. Centenaro, L. Vangelista, A. Zanella, M. Zorzi, Long-range commu-
nications in unlicensed bands: The rising stars in the IoT and smart city
scenarios, IEEE J. Wirel. Comm. 23 (5) (2016) 60–67.
[3] D. Patel, M. Won, Experimental study on low power wide area networks
for mobile internet of things, in: Proc. of VTC, Sydney, Australia, 2017,
pp. 1–5.
[4] U. Raza, P. Kulkarni, M. Sooriyabandara, Low power wide area networks:
An overview, IEEE J. Commun. Surv. Tuto. 19 (2) (2017) 855–873.
[5] A.M. Baharudin, W. Yan, Long-range wireless sensor networks for geo-
location tracking: Design and evaluation, in: Proc. of IES, Denpasar,
Indonesia, 2016, pp. 76–80.
[6] W. Guibene, J. Nowack, N. Chalikias, M. Kelly, Evaluation of LPWAN
technologies for smart cities: River monitoring use-case, in: Proc. of
WCNCW, San Francisco, CA, USA, 2017, pp. 17–22.
[7] O. Vondrous, Z. Kocur, T. Hegr, O. Slavicek, Performance evaluation of
IoT mesh networking technology in ISM frequency band, in: Proc. of ME,
Prague, Czech Republic, 2016, pp. 1–8.
[8] R. Sinha, Y. Wei, S. Hwang, A survey on LPWA technology: LoRa and
NB-IoT, J. ICT Expr. 3 (2017) 14–21.
[9] Sigfox world coverage available in: www.sigfox.com/en/coverage/.
[10] LoRa world coverage available in: www.lora-alliance.org/.
[11] F. Sforza, Communications system, 26 March 2013, US Patent US8406275
B2.
Please cite this article in press as: K. Mekki, et al., A comparative study of LPWAN technologies for large-scale IoT deployment, ICT Express (2018), https://doi.org/10.1016/j.icte.2017.12.005.
K. Mekki et al. / ICT Express ( ) – 7
[12] B. Reynders, W. Meert, S. Pollin, Range and coexistence analysis of long
range unlicensed communication, in: Proc. of ICT, Thessaloniki, Greece,
2016, pp. 51–56.
[13] K. Mikhaylov, T. Haenninen, Analysis of capacity and scalability of the
LoRa low power wide area network technology, in: Proc. of EWC, Oulu,
Finland, 2016, pp. 119–124.
[14] Y.E. Wang, X. Lin, A. Grovlen, Y. Sui, J. Bergman, A primer on 3GPP
narrowband internet of things, IEEE Commun. Mag. 55 (3) (2016) 117–
123.
[15] A. Adhikary, X. Lin, Y.-P.E. Wang, Performance evaluation of NB-IoT
coverage, in: Proc. of VTC-Fall, Montreal, QC, Canada, 2016, pp. 65–69.
[16] S.-M. Oh, J. Shin, An efficient small data transmission scheme in the
3GPP NB-IoT system, IEEE Commun. Lett. 21 (3) (2017) 660–663.
