ArticlePDF Available

Abstract and Figures

LoRa is a long-range, low-power, low-bitrate, wireless telecommunications system, promoted as an infrastructure solution for the Internet of Things: end-devices use LoRa across a single wireless hop to communicate to gateway(s), connected to the Internet and which act as transparent bridges and relay messages between these end-devices and a central network server. This paper provides an overview of LoRa and an in-depth analysis of its functional components. The physical and data link layer performance is evaluated by field tests and simulations. Based on the analysis and evaluations, some possible solutions for performance enhancements are proposed.
Content may be subject to copyright.
A Study of LoRa: Long Range & Low Power
Networks for the Internet of Things
Aloÿs Augustin1, Jiazi Yi 1,*, Thomas Clausen 1and William Mark Townsley 2
1École polytechnique Route de Saclay, 91128 Palaiseau, France; (A.A.); (T.C.)
2Cisco Paris Innovation and Research Laboratory (PIRL), 11 Rue Camille Desmoulins,
92782 Issy les Moulineaux, France;
*Correspondence:; Tel.: +33-177-578-085
Academic Editor: Dongkyun Kim
Received: 20 May 2016; Accepted: 1 September 2016 ; Published: 9 September 2016
LoRa is a long-range, low-power, low-bitrate, wireless telecommunications system,
promoted as an infrastructure solution for the Internet of Things: end-devices use LoRa across
a single wireless hop to communicate to gateway(s), connected to the Internet and which act as
transparent bridges and relay messages between these end-devices and a central network server.
This paper provides an overview of LoRa and an in-depth analysis of its functional components.
The physical and data link layer performance is evaluated by field tests and simulations. Based on
the analysis and evaluations, some possible solutions for performance enhancements are proposed.
Keywords: LoRa; Internet of Things; long range; low power
1. Introduction
The essential difference between “the Internet” and “the Internet of Things” (IoT) [
] is that in
the IoT, there is just “less of everything” available in a given device or network device: less memory,
less processing power, less bandwidth, etc.; and of course, less available energy. This is either because
“things” are battery driven and maximizing lifetime is a priority or because their number is expected to
be massive (it is estimated that there will be 50 billion connected devices by 2020 [
]). This drive to “do
more with less” leads to constraints that limit the applicability of traditional cellular networks, as well
as of technologies, such as WiFi, due to energy and scalability requirements.
Another range of protocols and technologies has emerged to fulfill the communication
requirements of the IoT: Low-Power Wide Area Networks (LPWAN). Colloquially speaking, an
LPWAN is supposed to be to the IoT what WiFi was to consumer networking: offering radio coverage
over a (very) large area by way of base stations and adapting transmission rates, transmission power,
modulation, duty cycles, etc., such that end-devices incur a very low energy consumption due to their
being connected.
LoRa (LoRa Alliance, is one such LPWAN protocol and the subject of
study for this paper. LoRa targets deployments where end-devices have limited energy (for example,
battery-powered), where end-devices do not need to transmit more than a few bytes at a time [
] and
where data traffic can be initiated either by the end-device (such as when the end-device is a sensor) or
by an external entity wishing to communicate with the end-device (such as when the end-device is an
actuator). The long-range and low-power nature of LoRa makes it an interesting candidate for smart
sensing technology in civil infrastructures (such as health monitoring, smart metering, environment
monitoring, etc.), as well as in industrial applications.
Sensors 2016,16, 1466; doi:10.3390/s16091466
Sensors 2016,16, 1466 2 of 18
1.1. Related Work
Different communication technologies aimed at low power, wireless IoT communication have
been proposed and deployed. As indicated above, these grossly fall within two categories:
Low power local area networks with a less than 1000-m range. This category includes
IEEE 802.15.4, IEEE P802.1ah, Bluetooth/LE, etc., which are applicable directly in short-range
personal area networks, in body area networks or, if organized in a mesh topology, also in
larger areas.
Low-power wide area networks, with a greater than 1000-m range, essentially low-power versions
of cellular networks, with each “cell” covering thousands of end-devices. This category includes
LoRaWAN, but also protocols, such as Sigfox, DASH7, etc.
This section provides a perspective on LoRaWAN by giving a brief overview of these related IoT
communication technologies.
1.1.1. IEEE802.15.4
IEEE 802.15.4 [
] is a standard specifying the physical layer and data link layer for Low-Rate
Wireless Personal Area Networks (LR-WPANs). Supporting three un-licensed frequency bands
(868 MHz, Europe; 928 MHz, North America; 2.4 GHz, worldwide), IEEE 802.15.4 can offer data
rates up to 250 kbit/s at a transmission range largely dependent on the environment; while for a clear
line-of-sight, up to 1000 m is possible; alas in most cases, the transmission range is measured in
tenths of meters. Built on top of the IEEE 802.15.4 physical and data link layers, ZigBee [
] offers
application-facing communications profiles and a network layer.
1.1.2. Bluetooth/LE
Released in 1999 by a consortium led by Ericsson, Nokia and Intel, Bluetooth v1.0 was initially
designed to, wirelessly, replace cables to connect devices typically used together, such as cell phones,
laptops, headsets, keyboards, etc., offering a lower data rate (1-Mbps raw data rate, max) and
a relatively short range (in theory, officially up to 100 m, at maximum transmission power, realistically,
5–10 m) while also a low power consumption.
Several revisions of Bluetooth later, Bluetooth 4.0 was completed in 2010. Fully compatible
with Bluetooth 1.0, this revision supports a higher data rate (24-Mbps raw data rate, based on WiFi)
and includes a “low energy” extension (called Bluetooth/LE or “Smart”). As compared with the
“non-LE version”, Bluetooth/LE provides rapid link establishment functions (simpler pairing) and
further trades off the data rate (approximately 200 kbps) for lower energy consumption, with the target
to run a wireless sensor for at least one year on a single coin cell (approximately 200 mAHr) [6].
1.1.3. IEEE 802.11 ah
] provides a wireless LAN standard that operates at sub-1-GHz license-exempt
bands. The work is conducted by the IEEE 802.11 ah Task Group (TGah). Compared to IEEE 802.11
(operating at 2.4 GHz and 5 GHz), 802.11 ah supports a longer transmission range up to 1 km at the
default transmission power of 200 mW. Depending on the bandwidth assigned, 802.11 ah can operate
at 4 Mbps or 7.8 Mbps. If the channel condition is good enough, 802.11 ah can provide a hundreds of
Mpbs data rate, thanks to the novel modulation and coding schemes brought from 802.11 ac.
1.1.4. Sigfox
Sigfox ( is a variation of the cellular system that enables remote devices
to connect to ab access point with Ultra Narrow Band (UNB). A proprietary technology, developed and
delivered by the French company Sigfox, no detailed public specification is available. Sigfox operates
on the 868-MHz frequency band, with the spectrum divided into 400 channels of 100 Hz [
Sensors 2016,16, 1466 3 of 18
Each end-device can send up to 140 messages per day, with a payload size of 12 octets, at a data rate up
to 100 bps. Sigfox claims that each access point can handle up to a million end-devices, with a coverage
area of 30–50 km in rural areas and 3–10 km in urban areas. Sigfox’s claim to being a low power
technology stems, in no small part, from end-devices being heavily duty-cycled due to an assumption
of the nature of the data traffic patterns in the IoT: when an end-device has a message to send, the Sigfox
interface circuitry wakes up, and the message is transmitted “uplink”, from the end-device; then, the
end-device listens for a short duration in case there are data being sent “downlink”, to the end-device.
In other words, downlink traffic is supported by the end-device actively polling, which makes Sigfox
an interesting choice for data acquisition, but perhaps less so for command-and-control scenarios.
1.1.5. DASH7
] is a wireless sensor and actuator full Open Systems Interconnection (OSI) stack
protocol that operates in the 433-MHz, 868-MHz and 915-MHz unlicensed ISM band/SRD band.
It originates from the ISO 18000-7 standard [
] for active RFID, intended by the U.S. Department of
Defense for container inventory. DASH7 inherits from ISO/IEC 18000-7 the default parameters of the
active air interface communication at 433 MHz, an asynchronous MAC and a presentation layer using
highly structured data elements. Furthermore, DASH7 extends and defines the protocol stack from the
physical layer up to the application layer.
DASH7 aims at providing communication in the range of up to 2 km, low latency, mobility
support, multi-layer battery life, AES 128-bit shared key encryption support and a data rate up to
167 kbit/s.
In [
], a more detailed survey of different technologies, including 3GPP LTE Rel-13, Nokia’s
narrow-band LTE-M, Neul/Huawei’s narrow-band proposal, Sigfox, etc., is provided.
1.2. Statement of Purpose
There have been a few articles related to LoRa in the literature. In [
], different long-range
technologies, including LoRa, are compared. Petajajarvi et al. [
] studied the coverage of LoRa and
proposed a channel attenuation model. In [
], the authors analyzed the LoRa capacity and proposed
LoRaBlink to support multi-hop communications.
In complementing the work of these articles, the goal of this paper is three-fold: (i) given the
semi-proprietary nature of LoRA (parts of the protocol are well documented; other parts are not), to
provide an overview and functional description of LoRa and to present as much information as could be
(experimentally and otherwise) gathered; (ii) to independently provide a quantification and evaluation
of the performance of LoRA and of LoRaWAN, especially the spreading factor; and (iii) based on the
analysis and performance evaluation, to propose possible solutions for performance enhancement.
The remainder of this paper is organized as follows: Section 2provides a functional overview
of LoRa, followed by Section 3, which describes and analyzes the LoRa physical layer in detail
and provides experimental performance studies hereof. Following, the LoRaWAN MAC protocol
is described in Section 4, with Section 5presenting the evaluation hereof for LoRaWAN. Section 6
concludes this paper.
2. LoRa Overview
This section gives an overview of the LoRa protocol stack and basic network architecture.
2.1. LoRa Protocol Stack
LoRa, which stands for “Long Range”, is a long-range wireless communications system, promoted
by the LoRa Alliance. This system aims at being usable in long-lived battery-powered devices, where
the energy consumption is of paramount importance. LoRa can commonly refer to two distinct layers:
(i) a physical layer using the Chirp Spread Spectrum (CSS) [
] radio modulation technique; and
Sensors 2016,16, 1466 4 of 18
(ii) a MAC layer protocol (LoRaWAN), although the LoRa communications system also implies a
specific access network architecture.
The LoRa physical layer, developed by Semtech, allows for long-range, low-power and
low-throughput communications. It operates on the 433-, 868- or 915-MHz ISM bands, depending
on the region in which it is deployed. The payload of each transmission can range from 2–255 octets,
and the data rate can reach up to 50 Kbps when channel aggregation is employed. The modulation
technique is a proprietary technology from Semtech.
LoRaWAN provides a medium access control mechanism, enabling many end-devices to
communicate with a gateway using the LoRa modulation. While the LoRa modulation is proprietary,
the LoRaWAN is an open standard being developed by the LoRa Alliance.
2.2. LoRa Network Architecture
A typical LoRa network is “a star-of-stars topology”, which includes three different types of
devices, as shown in Figure 1.
LoRa gateway
LoRa Network server
LoRa gateway
End device End device End device
IP connection
LoRa connection
Figure 1. LoRa network architecture.
The basic architecture of a LoRaWAN network is as follows: end-devices communicate with
gateways using LoRa with LoRaWAN. Gateways forward raw LoRaWAN frames from devices to
a network server over a backhaul interface with a higher throughput, typically Ethernet or 3G.
Consequently, gateways are only bidirectional relays, or protocol converters, with the network server
being responsible for decoding the packets sent by the devices and generating the packets that should
be sent back to the devices. There are three classes of LoRa end-devices, which differ only with regards
to the downlink scheduling.
3. The LoRa Physical Layer
The LoRa modulation is a Semtech proprietary technology and is as such not fully open.
This section presents an analysis (of the parts of LoRa that are open) and an experimental evaluation
(of the proprietary parts of LoRa) with the purpose of understanding if the advertised performance of
LoRa is observed in practice.
Sensors 2016,16, 1466 5 of 18
3.1. Overview of the Physical Layer
LoRa is a chirp spread spectrum modulation [
], which uses frequency chirps with a linear
variation of frequency over time in order to encode information. Because of the linearity of the chirp
pulses, frequency offsets between the receiver and the transmitter are equivalent to timing offsets, easily
eliminated in the decoder. This also makes this modulation immune to the Doppler effect, equivalent
to a frequency offset. The frequency offset between the transmitter and the receiver can reach 20%
of the bandwidth without impacting decoding performance [
]. This helps with reducing the price
of LoRa transmitters, as the crystals embedded in the transmitters do not need to be manufactured
to extreme accuracy. LoRa receivers are able to lock on to the frequency chirps received, offering a
sensitivity of the order of 130 dBm [19,20].
As the LoRa symbol duration is longer than the typical bursts of AMinterference generated by
Frequency Hopping Spread Spectrum (FHSS) systems, errors generated by such interference are easily
corrected through Forward Error-correction Codes (FECs). The typical out-of-channel selectivity (the
maximum ratio of power between an interferer in a neighboring band and the LoRa signal) and
co-channel rejection (the maximal ratio of power between an interferer in the same channel and the
LoRa signal) of LoRa receivers is respectively 90 dB and 20 dB [
]. This outperforms traditional
modulation schemes, such as Frequency-Shift Keying (FSK), and makes LoRa well suited to low-power
and long-range transmissions.
3.2. Parameters of the Physical Layer
Several parameters are available for the customization of the LoRa modulation: Bandwidth (
Spreading Factor (
) and Code Rate (
). LoRa uses an unconventional definition of the spreading
factor as the logarithm, in base 2, of the number of chirps per symbol. For the sake of simplicity, this
article will stick to this definition. Theses parameters influence the effective bitrate of the modulation,
its resistance to interference noise and its ease of decoding.
The bandwidth is the most important parameter of the LoRa modulation. A LoRa symbol is
composed of 2
chirps, which cover the entire frequency band. It starts with a series of upward chirps.
When the maximum frequency of the band is reached, the frequency wraps around, and the increase in
frequency starts again from the minimum frequency. Figure 2gives an example of a LoRa transmission
in the frequency variation over time. The position of this discontinuity in frequency is what encodes
the information transmitted. As there are 2
chirps in a symbol, a symbol can effectively encode
bits of information.
Figure 2. Frequency variation over time of a sample signal emitted by a LoRa transmitter. Data taken
from [21]. fcis the central frequency of the channel, and BW is the bandwidth.
In LoRa, the chirp rate depends only on the bandwidth: the chirp rate is equal to the bandwidth
(one chirp per second per Hertz of bandwidth). This has several consequences on the modulation:
an increase of one of the spreading factor will divide the frequency span of a chirp by two (as 2
cover the whole bandwidth) and multiply the duration of a symbol by two, also. It will not, however,
divide the bit rate by two, as one more bit will be transmitted in each symbol. Moreover, the symbol rate
and the bit rate at a given spreading factor are proportional to the frequency bandwidth, so a doubling
Sensors 2016,16, 1466 6 of 18
of the bandwidth will effectively double the transmission rate. This is translated in Equation (1), which
links the duration of a symbol (TS) to the bandwidth and the spreading factor.
BW (1)
Moreover, LoRa includes a forward error correction code. The code rate (
) equals 4
n∈ {
1, 2, 3, 4
. Taking this into account, as well as the fact that
bits of information are
transmitted per symbol, the Equation (2) allows one to compute the useful bit rate (Rb).
2SF ×CR (2)
For example, a setting with BW =125 kHz, SF =7, CR =4/5 gives a bit rate of Rb= 5.5 kbps.
These parameters also influence decoder sensitivity. Generally speaking, an increase of bandwidth
lowers the receiver sensitivity, whereas an increase of the spreading factor increases the receiver
sensitivity. Decreasing the code rate helps reduce the Packet Error Rate (PER) in the presence of
short bursts of interference, i.e., a packet transmitted with a code rate of 4
8 will be more tolerant to
interference than a signal transmitted with a code rate of 4
5. The figures in Table 1, taken from the
SX1276 datasheet, are given as an indication.
Table 1.
Semtech SX1276 LoRa receiver sensitivity in dBm at different bandwidths and spreading
factors, taken from [19].
SF 7 8 9 10 11 12
125 kHz 123 126 129 132 133 136
250 kHz 120 123 125 128 130 133
500 kHz 116 119 122 125 128 130
Another parameter of the LoRa modulation, which is implemented in Semtech’s transceivers, is
the low data rate optimization. This parameter is mandatory in LoRa when using spreading factors
of 11 and 12 with a bandwidth of 125 kHz or lower. The effect of this parameter is not documented;
however, Equation (3) shows that it reduces the number of bits transmitted per symbol by two.
3.3. Physical Frame Format
Although the LoRa modulation can be used to transmit arbitrary frames, a physical frame format
is specified and implemented in Semtech’s transmitters and receivers. The bandwidth and spreading
factor are constant for a frame.
A LoRa frame begins with a preamble. The preamble starts with a sequence of constant upchirps
that cover the whole frequency band. The last two upchirps encode the sync word. The sync word is a
one-byte value that is used to differentiate LoRa networks that use the same frequency bands. A device
configured with a given sync word will stop listening to a transmission if the decoded sync word
does not match its configuration. The sync word is followed by two and a quarter downchirps, for a
duration of 2.25 symbols. The total duration of this preamble can be configured between 10.25 and
65,539.25 symbols. The structure of the preamble can be seen in Figure 2.
After the preamble, there is an optional header. When it is present, this header is transmitted
with a code rate of 4/8. This indicates the size of the payload (in bytes), the code rate used for the
end of the transmission and whether or not a 16-bit CRCfor the payload is present at the end of the
frame. The header also includes a CRC to allow the receiver to discard packets with invalid headers.
The payload size is stored using one byte, limiting the size of the payload to 255 bytes. The header is
optional to allow disabling it in situations where it is not necessary, for instance when the payload
length, coding rate and CRC presence are known in advance.
Sensors 2016,16, 1466 7 of 18
The payload is sent after the header, and at the end of the frame is the optional CRC. A schematic
summarizing the frame format can be seen in Figure 3.
Figure 3. Structure of a LoRa frame. n∈ {1..4}.
Equation (3), derived from Semtech’s datasheets [
], gives the number of symbols required
to transmit a payload
, as a function of all of these parameters. This number should be added to
the number of symbols of the preamble, in order to compute the total size of the packet in symbols.
In this equation,
is the payload size in bytes,
is 16 if the CRC is enabled and zero otherwise,
is 20 when the header is enabled and zero otherwise and
is two when the low data rate
optimization is enabled and zero otherwise. This equation also shows that the minimum size of a
packet is eight symbols.
ns=8+max 8PL 4SF +8+CRC +H
4×(SF DE)×4
CR , 0(3)
3.4. Performance Evaluation
To verify whether the specified performance of LoRa receivers is reached in practice, a LoRa
testbed is built. The Freescale KRDM-KL25Z development board with Semtech SX1276 MBED shield
(Figure 4a) is used as the end-device, and a Cisco 910 industrial router is used as the gateway (Figure 4b).
The gateway is connected to the network server provided by Thingpark (
through Ethernet, so that the packet received can be monitored on the server side.
(a) (b)
Figure 4. The LoRa testbed. (a) The LoRa end-device; (b) the LoRa gateway.
3.4.1. Receiver Sensitivity
As there are many models and evaluations of the propagation of radio signals at the frequencies
used by LoRa in various environments [
], this experiment is focused on checking the decoding
performance of LoRa receivers.
To this end, around 10,000 packets were sent from a LoRa device to the gateway, and the Received
Signal Strength Indicators (RSSI) of received packets were recorded while moving the end-device.
Sensors 2016,16, 1466 8 of 18
The gateway was placed indoors, and the device was outdoors, in an urban environment. All packets
were sent with a bandwidth of 125 kHz and a code rate of 4/5. The transmit power of the device
was set to the minimum (2 dBm, with a 3-dBi antenna) in order to limit the distance to cover before
reaching low RSSIs. The order of magnitude of the distance between the end-device and the gateway
at which packets started to get lost was 100 m. The minimal observed RSSIs are depicted in Figure 5.
Figure 5. Minimal observed RSSIs with different spreading factors.
These measured results are slightly above the specified values, and the expected decrease with
the increase of the spreading factor is not observed. However, the packets achieving the lowest RSSIs
were also received with a high SINR, close to 20 dB. This is likely due to the gateway being indoors,
leading to additional shadowing.
It should be noted that the observed RSSIs are already 6 dB lower than the specified RSSIs when
using FSK [19].
3.4.2. Network Coverage
This experiment aims at testing the network coverage of LoRa. Tests were conducted in a
suburb of Paris, with mainly low-rise residential dwellings. The temperature was 15
C, and the
ambient humidity was 55%. The gateway was located on the second floor of a house, outside the
window. Five different test points were chosen, with the distance to the gateway as shown in Figure 6.
The end-device was in a car during the tests.
The transmission power of the end-device was set to 14 dBm, which is the default value as
specified by [
]. To test the performance of different spreading factors, the packet acknowledgment
and retransmission was turned off. The link check was also disabled so that the spreading factor will
not change even if there is packet loss; by default, LoRa will adapt the spreading factor according to
the link quality. Spreading factors of 7, 9 and 12 were chosen for the tests.
Figure 7shows the packet delivery ratio of different spreading factors with various distances.
About 100 packets are transmitted to the network server in each test with a sequence number.
The higher spreading factors have better coverage, as discussed in Section 3.2: for a spreading factor
of 12, more than 80% of packets were received at Point D (2800 m), while no packet was received when
using a spreading factor of seven. It is worth noting that the gateway was located in the second floor,
which was about 5 m above the ground (normally, such a base station would be located at a higher
altitude to achieve better coverage), and the test Point D was right behind a building of seven floors.
The high delivery ratio using the high spreading factor has the cost of a much lower bit rate, as shown
in Equation (2). On the other hand, the network coverage with low spreading factors is much lower.
Sensors 2016,16, 1466 9 of 18
2300m 2800m
Figure 6. Map of LoRa field test.
SP31 2
Figure 7. Packet delivery ratio of the LoRa field test.
It is important to note that the purpose of the tests above is to test the coverage of the LoRa
physical layer using different spreading factors. In a real LoRa network with the LoRaWAN protocol,
the end-devices are able to automatically increase the spreading factor if the transmission with the
lower spreading factor fails. Furthermore, retransmission is also used if necessary. Therefore, in a
network with LoRaWAN, a higher delivery ratio can be achieved.
4. The LoRaWAN Protocol
LoRaWAN is a MAC protocol, built to use the LoRa physical layer. It is designed mainly for sensor
networks, wherein sensors exchange packets with the server with a low data rate and relatively long
time intervals (one transmission per hour or even days). This section describes the LoRaWAN V1.0
specification [23], as released in January 2015.
4.1. Components of a LoRaWAN Network
Several components of the network are defined in the LoRaWAN specification and are required to
form a LoRaWAN network: end-devices, gateways (i.e., base stations) and the network server.
End-device: the low-power consumption sensors that communicate with gateways using LoRa.
Sensors 2016,16, 1466 10 of 18
Gateway: the intermediate devices that forward packets coming from end-devices to a network
server over an IP backhaul interface allowing a bigger throughput, such as Ethernet or 3G.
There can be multiple gateways in a LoRa deployment, and the same data packet can be received
(and forwarded) by more than one gateway.
Network server: responsible for de-duplicating and decoding the packets sent by the devices and
generating the packets that should be sent back to the devices.
Unlike traditional cellular networks, the end-devices are not associated with a particular gateway
in order to have access to the network. The gateways serve simply as a link layer relay and forward
the packet received from the end-devices to the network server after adding information regarding the
reception quality. Thus, an end-device is associated with a network server, which is responsible
for detecting duplicate packets, choosing the appropriate gateway for sending a reply (if any),
consequently for sending back packets to the end-devices. Logically, gateways are transparent to
the end-devices.
LoRaWAN has three different classes of end-devices to address the various needs of applications:
Class A, bi-directional: Class A end-devices can schedule an uplink transmission based on their
own needs, with a small jitter (random variation before transmission). This class of devices
allows bi-directional communications, whereby each uplink transmission is followed by two short
downlink receive windows. Downlink transmission from the server at any other time has to wait
until the next uplink transmission occurs. Class A devices have the lowest power consumption,
but also offer less flexibility on downlink transmissions.
Class B, bi-directional with scheduled receive slots: Class B end-devices open extra receive
windows at scheduled times. A synchronized beacon from the gateway is thus required, so that
the network server is able to know when the end-device is listening.
Class C, bi-directional with maximal receive slots: Class C end-devices have almost continuous
receive windows. They thus have maximum power consumption.
It should be noted that LoRaWAN does not enable device-to-device communications: packets can
only be transmitted from an end-device to the network server, or vice-versa. Device-to-device
communication, if required, must thus be sling-shot through the network server (and consequently, by
way of two gateway transmissions).
The LoRaWAN specification states that LoRaWAN networks should use ISM frequency bands.
These bands are subject to regulations regarding the maximum transmission power and the duty cycle.
These duty cycle limitations translate into delays between the successive frames sent by a device. If the
limitation is at 1%, the device will have to wait 100-times the duration of the last frame before sending
again in the same channel.
4.2. LoRaWAN Message Format
LoRaWAN uses the physical frame format described in Section 3.3. The header and CRC are
mandatory for uplink messages, which makes it impossible to use a spreading factor of six in
LoRaWAN. Downlink messages have the header, but not the CRC. The code rate that should be
used is not specified and neither is when the end-devices should use the low data rate optimization.
The message format is detailed in Figure 8.DevAddr is the short address of the device. FPort is
a multiplexing port field. The value zero means that the payload contains only MAC commands.
When this is the case, the FOptsLen field must be zero. FCnt is a frame counter. MIC is a cryptographic
message integrity code, computed over the fields MHDR,FHDR,FPort and the encrypted FRMPayload.
MType is the message type, indicating among other things whether it is an uplink or a downlink
message and whether or not it is a confirmed message. Acknowledgments are requested for confirmed
messages. Major is the LoRaWAN version; currently, only a value of zero is valid. ADR and ADRAckReq
control the data rate adaptation mechanism by the network server. ACK acknowledges the last received
Sensors 2016,16, 1466 11 of 18
frame. FPending indicates that the network server has additional data to send and that the end-device
should send another frame as soon as possible so that it opens receive windows. FOptsLen is the length
of the FOpts field in bytes. FOpts is used to piggyback MAC commands on a data message. CID is the
MAC command identifier, and Args are the optional arguments of the command. FRMPayload is the
payload, which is encrypted using AES with a key length of 128 bits. The minimal size of the MAC
header is 13 bytes; its maximal size is 28 bytes. Knowing this, it is possible to compute the maximum
channel capacity available for application data payloads with given modulation parameters thanks to
Equations (1) and (3). As packets are sent from a device to the network server and vice versa, there is
no destination address on uplink packets, and there is no source address on downlink packets.
Figure 8. LoRaWAN frame format. The sizes of the fields are in bits.
4.3. End-Device Setup
In order to participate in a LoRaWAN network, an end-device must be activated.
LoRaWAN provided two ways to activate an end-device: Over-The-Air Activation (OTAA) and
Activation By Personalization (ABP).
The activation process should give the following information to an end-device:
End-device address (DevAddr): A 32-bit identifier of the end-device. Seven bits are used as the
network identifier, and 25 bits are used as the network address of the end-device.
Application identifier (AppEUI): A global application ID in the IEEE EUI64 address space that
uniquely identifies the owner of the end-device.
Network session key (NwkSKey): A key used by the network server and the end-device to calculate
and verify the message integrity code of all data messages to ensure data integrity.
Application session key (AppSKey): A key used by the network server and end-device to encrypt
and decrypt the payload field of data messages.
For OTAA, a join procedure with a join-request and a join-accept message exchange is used for
each new session. Based on the join-accept message, the end-devices are able to obtain the new session
keys (NwkSkey and AppSKey). For the ABP, the two session keys are directly stored into the end-devices.
4.4. LoRaWAN MAC Commands
LoRaWAN defines many MAC commands that allow customizing end-device parameters [
One of them, LinkCheckReq, can be sent by an end-device to test its connectivity. All of the others are sent
by the network server. These commands can control the data rate and output power used by the device,
as well as the number of times each unconfirmed packet should be sent (LinkADRReq), the global duty
cycle of the device (DutyCycleReq), changing parameters of the receive windows (RXTimingSetupReq,
RXParamSetupReq) and changing the channels used by the device (NewChannelReq). One command is
used to query the battery level and reception quality of a device (DevStatusReq).
Sensors 2016,16, 1466 12 of 18
5. LoRaWAN Analysis
This section analyzes and discusses the performance of LoRaWAN by way of experiments and
simulations. As in the previous section, all of this study is based on [23].
5.1. Single Device Maximal throughput and MTU
The goal of this experiment is to evaluate the maximal throughput that a single device can obtain.
This depends more on the physical layer than on the MAC protocol, but it gives an idea of what is
possible when using LoRaWAN. The experiment was conducted by having a device send data as soon
as the channel limitations and the protocol allow it. Tests were conducted with six channels of 125 kHz
and using spreading factors from 7–12. No MAC commands were sent, so the size of the MAC header
was always 13 bytes. The results, depending on the payload size, are visible in Figure 9, which are
measured over about 100 packets transmitted in each test. Fifty-one bytes are the maximum payload
size allowed by the implementation used for the tests.
! # + *$ ** *(
,-./,0.1 23/ 456 5217 892.:;:<
:6/.,=>? 01@,A24/
Figure 9. Maximum throughput attained by a single device using LoRaWAN.
This experiment revealed that at low packet sizes, the limiting factor was not the channel duty
cycle limitations, as could have been expected, but the duration of the receive windows. Indeed, the
device has to wait for the two downlink receive windows following the transmission to be over before
sending another packet. However, this situation is not the use case LoRaWAN was designed for:
the goal of LoRaWAN is rather to manage large quantities of devices that send a few bytes of data
from time to time.
In the tests above, the MAC header is always 13 bytes. However, in practice, the LoRaWAN
header can be a variable size between 13 and 28 bytes. Moreover, the maximum size of the frame
depends on the data rate used [
], and LoRaWAN does not have a mechanism to split large payloads
over multiple frames. As of the current specification, the application above LoRaWAN has no way of
knowing what the maximal size of the packet that it will be able to send in the next transmission is,
which might be problematic. A conservative approach is to never try to send more than the smallest
maximum payload size, which is 36 bytes, but this results in a loss of capacity if a large amount of data
has to be sent, as well as lower throughput, as shown in the results in Figure 9. This is relatively easy
to address in a future LoRaWAN specification revision, either by adding a fragmentation mechanism
or by informing the upper layer of the MTU from MAC protocol.
5.2. Total Capacity and Channel Load
The total capacity of the network is not only related to payload size. As two transmissions on the
same frequency, but at different spreading factors, can be decoded simultaneously, in what follows, a
logical channel is defined by a pair (frequency band, spreading factor).
Sensors 2016,16, 1466 13 of 18
The total transmission capacity of a LoRaWAN network is the sum of the capacities of all of the
logical channels. In a 125-kHz frequency band, there are six possible spreading factors (from 7 to 12),
which brings the total capacity of a 125-kHz channel to 12,025 bps.
In the EU frequency band, the set of mandatory channels contains three 125-kHz channels [
which make the minimum total capacity of the network 36 kbps. Networks operators are free to add
more channels (sent to the devices using NewChannelReq commands), thus increasing the capacity of
the network.
As the transmission bit rate is dependent on the spreading factor, not all logical channels have the
same capacity. In what follows, the load for a logical channel is defined by the time average of the
numbers of LoRa devices trying to send data. This coincides with the natural definition of the load: in
optimal conditions, i.e., with a perfect synchronization of the devices, a load of one can be reached,
saturating the channel.
5.3. Estimation of the Collision Rate
As of the current specification, the devices and the gateways can transmit at any time. There is
no listen-before-talk or CSMAmechanism. This makes LoRaWAN very similar to ALOHA [
], but
contrary to ALOHA with a variable packet length.
Because of the legal duty-cycle limitations of 1% in the EU region where this analysis took place,
100 devices would have been needed to emulate a load of one, and this number would have grown
proportionally to the maximum link load we wanted to test. As there were not this many devices on
hand, simulations are used to evaluate LoRaWAN’s behavior under load.
A simulator was built to simulate the random process of packet emissions. Five-hundred-thousand
packets were simulated for each data point. If the transmission time of two packets overlaps, we
consider that a collision happens and that none of the two packets reaches the gateway. The collision
rate is the number of packets that collided, divided by the total number of packets sent during
the simulation. The channel capacity usage is computed as the amount of data that is successfully
transferred during the simulation, divided by the theoretical maximum amount of data that could
have been sent in the channel, which is the channel capacity multiplied by the simulation duration.
The channel load is as defined in the previous Section 5.2, or equivalently, the sum of the duration of
all of the packets sent during the simulation, divided by the duration of the simulation.
The duration of the packets for the different payload sizes was computed using Semtech’s LoRa
Calculator, for a spreading factor of seven, a bandwidth of 125 kHz, a code rate of 5/4 and six symbols
in the preamble.
Assuming the packet arrivals are following a Poisson law and a uniform distribution of the
payloads lengths between one and 51 bytes, the expected capacity usage and collision rate depending
on the load for one logical channel can be plotted. The result is shown in Figure 10.
The variable packet length does not greatly impact the performance of LoRaWAN, and all said
and done, the observed behavior is very close to that of pure ALOHA. The maximum capacity usage
is 18% of the channel capacity and is reached for a link load of 0.48. However, at this load, around 60%
of the packets transmitted are dropped because of collisions.
This may be an issue, because if the devices are not using confirmed messages, some messages
will be lost (and increasing the number of times each message is sent by the devices is a bad solution,
as it will increase the load on the link), and if the devices are using confirmed messages, they will have
to retransmit most packets several times, which will in addition impact the battery life of the devices.
Sensors 2016,16, 1466 14 of 18
Figure 10.
Link capacity usage and packet collision rate for a LoRaWAN network and compared to an
ALOHA network. The load is as defined in Section 5.2.
LoRaWAN confirmed messages sent by the devices must be acknowledged by a packet sent
during one of the two receive windows following the transmission, while confirmed messages sent
by the gateway will be acknowledged during the next uplink transmission. The acknowledgment is
only a flag in the packet header, and the setting of this flag acknowledges the last message received.
As such, when using confirmed messages, a new packet should not be sent before the acknowledgment
of the previous packet was received; otherwise, it will be impossible to know to which packet the next
acknowledgment will be referring.
The drawback of this mechanism is that a confirmed message requires two successive
transmissions in order to be successful, thus increasing the collision probability with other messages
and the number of retransmissions needed. As above, the probability of success and the link
capacity usage when end-devices are sending confirmed messages can be plotted. For this simulation,
we consider that the gateway does not send MAC commands to the device, so the acknowledgment
message is always using a 13-byte MAC header and no payload. We also take these messages into
account in the computation of the load, i.e., when the sum of the durations of all of the messages
and their acknowledgments is equal to the duration of the simulation, the value of the load is one.
The result is shown in Figure 11.
Figure 11.
Link capacity usage and packet collision rate for a LoRaWAN network when using confirmed
messages. The load is as defined in Section 5.2.
Sensors 2016,16, 1466 15 of 18
As expected, the success rate is significantly lower than without confirmed messages.
However, this is a relatively efficient way of implementing this functionality, because two successful
transmissions are necessary anyway.
The results show that LoRaWAN is extremely sensitive to the channel load, similar to ALOHA.
The solution implemented by usual network protocols, such as 802.11 or cellular networks, to help
mitigate this problem is CSMA [
]. In order to ensure the scalability of LoRaWAN, it could be
interesting to study the feasibility of the implementation of a CSMA mechanism into LoRaWAN.
A possible issue is the duty cycle limitation that applies to the gateway, and that would prevent it from
sending messages too often; another is the potential non-transitivity of the channel (i.e., an end-device
may or may not be able to “carrier sense” if another end-device is transmitting to the same gateway).
If the current architecture is kept, the CSMA mechanism would have to be controlled by the network
server, which would put even more load on it. Alas, a CSMA mechanism could also remove
the risk of collision of the acknowledgment for confirmed messages, by making it happen during
a contention-free period.
The current LoRaWAN specification does not have any means to enforce quality of service, and
thus, it should not be used for critical applications or applications where the delay between the first
time at which the device tries to send a message and the time at which it is received is important.
Adjusting the number of times a device sends its packets may increase the chance of these packets
going through, but it does so at the expense of more collisions with transmissions from other nodes
and does not provide any hard guarantee.
LoRaWAN currently uses ISM bands, which have the advantage of being free and not requiring
a license. However, these bands are more and more used by LoRaWAN’s competitors. Even if LoRa
is very resistant to interferences, these bands have a finite capacity, and it is not guaranteed that the
capacity of this band is sufficient. Moreover, it is perfectly legal for a malicious individual to emit
random LoRa symbols, which will jam LoRa transmissions. Using a proprietary frequency band
would have the advantage to remove most interferences, as well as remove the duty cycle cap, possibly
making the implementation of a CSMA mechanism easier.
5.4. The Network Server Role
LoRaWAN specifies the behavior of the devices, but not the behavior of the network server.
As shown in Section 5.3, it is important to keep the load on the network low, and the network server
has to enforce this by sending MAC commands to the devices. However, as this is not part of the
specification and as there is no open source reference implementation (as of the writing of this article),
a correct behavior of the network server is hard to be evaluated.
The network server can easily degrade the performance of the network. For instance, it can
use the LinkADRReq command to configure the number of times a device will send each data frame.
This parameter is advertised as a way to control the quality of service for a device. Setting this
parameter to more than one will increase the load on the network, increasing the amount of collisions,
and thus, should be done very cautiously.
Moreover, LoRaWAN networks are advertised to be able to handle millions of devices.
The network server will be responsible for the optimization of all of these nodes. Even if the event rate
in sensor networks is significantly lower than in traditional networks, the performance of the network
server should be carefully evaluated by the network operators, to ensure the scaling of the network.
5.5. The Gateway Role
The current specification states that the gateway is only a relay. This is linked to the fact that the
packets sent by the devices have no destination address (which saves a few bytes) and that there is no
association between a device and a gateway. Indeed, as several gateways can receive the same message
from a device, only one of them should reply to it. It falls back to the network server to choose the
best gateway.
Sensors 2016,16, 1466 16 of 18
The only task that should be handled by the gateways is the timing of the downlink messages.
This timing should be accurate so that the device receives the message in its receive window. It is not
specified whether the gateways receive a message to send from the server along with the time at which
it should be sent or if the gateway sends the message received from the server as soon as it receives it,
and it is unclear which solution is implemented in existing gateways. As the round trip time of the
backhaul interface of the gateways cannot be controlled, the first solution should be implemented.
It would also allow one to synchronize the transmissions of the different gateways, avoiding collisions
between them.
In the current specification, each gateway is dedicated to a specific network server, as shown in
Figure 1. This means both the gateways and the data collected are “owned” by the entity that runs
the only network server. In the future, it would be interesting to extend the function of the gateways
so that they can forward the packet to specific network servers, as shown in Figure 12. This may
effectively reduce the expense of devices and network deployment.
LoRa gateway
LoRa Network server
Net A
LoRa gateway
End device
Net A
End device
Net A
IP connection
LoRa connection
End device
Net B
End device
Net B
LoRa Network server
Net B
Figure 12.
An example of shared gateways in LoRa. The gateways can forward the packet to different
network servers.
6. Conclusions
LoRa is a long-range and low-power telecommunication systems for the “Internet of Things”.
The physical layer uses the LoRa modulation, a proprietary technology with a MAC protocol.
LoRaWAN is an open standard with the specification available free of charge [23].
This paper gives a comprehensive analysis of the LoRa modulation, including the data rate,
frame format, spreading factor, receiver sensitivity, etc. A testbed has been built, to experimentally
study the network performance, documented in this paper . The results show that LoRa modulation,
thanks to the chirp spread spectrum modulation and high receiver sensitivity, offers good resistance
to interference. Field tests show that LoRa can offer satisfactory network coverage up to 3 km in a
suburban area with dense residential dwellings. The spreading factor has significant impact on the
network coverage, as does the data rate. LoRa is thus well suited to low-power, low-throughput and
long-range networks.
This paper has also shown that LoRaWAN is an LPWAN protocol very similar to ALOHA.
Its performance thus degrades quickly when the load on the link increases.
This work was supported in part by the Cisco-Polytechnique Chaire “Internet of Everything”
( The authors would like to particularly thank Patrick Grossetête,
Cisco Paris Research and Innovation Laboratory (PIRL), France, for supplying a LoRA gateway and for providing
access to a LoRa network server.
Sensors 2016,16, 1466 17 of 18
Author Contributions:
Aloÿs Augustin, Jiazi Yi, Thomas Clausen and Mark Townsley conceived, designed
and performed the experiments; Aloÿs Augustin, Jiazi Yi, Thomas Clausen and Mark Townsley contributed
materials/analysis tools; Aloÿs Augustin, Jiazi Yi, Thomas Clausen and Mark Townsley wrote the paper.
Conflicts of Interest:
The authors declare no conflict of interest. The founding sponsors had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the
decision to publish the results.
Gubbi, J.; Buyya, R.; Marusic, S.; Palaniswamia, M. Internet of Things (IoT): A vision, architectural elements,
and future directions. Future Gener. Comput. Syst. 2013,29, 1645–1660.
Evans, D. The Internet of Things: How the Next Evolution of the Internet is Changing Everything; Cisco Internet
Business Solutions Group: San Jose, CA, USA, 2011.
LoRa Alliance. White Paper: A Technical Overview of Lora and Lorawan; The LoRa Alliance: San Ramon, CA,
USA, 2015.
IEEE 802 Working Group and Others. IEEE Standard for Local and Metropolitan Area Networks—Part 15.4:
Low-Rate Wireless Personal Area Networks (LR-WPANs); IEEE Std 802.15.4-2011: New York, NY, USA, 2012.
Baronti, P.; Pillai, P.; Chook, V.W.; Chessa, S.; Gotta, A.; Hu, Y.F. Wireless sensor networks: A survey on the
state of the art and the 802.15. 4 and ZigBee standards. Comput. Commun. 2007,30, 1655–1695.
6. Want, R.; Schilit, B.; Laskowski, D. Bluetooth le finds its niche. IEEE Pervasive Comput. 2013,12, 12–16.
Khorov, E.; Lyakhov, A.; Krotov, A.; Guschin, A. A survey on ieee 802.11 ah: An enabling networking
technology for smart cities. Comput. Commun. 2015,58, 53–69.
Adame, T.; Bel, A.; Bellalta, B.; Barcelo, J.; Oliver, M. IEEE 802.11 ah: the wifi approach for M2M
communications. IEEE Wirel. Commun. 2014,21, 144–152.
Margelis, G.; Piechocki, R.; Kaleshi, D.; Thomas, P. Low throughput networks for the IoT: Lessons learned
from industrial implementations. In Proceedings of the 2015 IEEE 2nd World Forum on Internet of Things
(WF-IoT), Milan, Italy, 14–16 december 2015; pp. 181–186.
Weyn, M.; Ergeerts, G.; Wante, L.; Vercauteren, C.; Hellinckx, P. Survey of the DASH7 alliance protocol for
433 MHz wireless sensor communication. Int. J. Distrib. Sens. Netw. 2013,9, 870430.
ISO-IEC. ISO/IEC 18000-7:2014 Information Technology—Radio Frequency Identification for Item
Management—Part 7: Parameters for Active Air Interface Communications at 433 MHz; ISO: Geneva,
Switherland, 2014.
Guibene, W.; Nolan, K.E.; Kelly, M.Y. Survey on clean slate cellular-iot standard proposals. In Proceedings of
the 2015 IEEE International Conference on Computer and Information Technology; Ubiquitous Computing
and Communications; Dependable, Autonomic and Secure Computing; Pervasive Intelligence and
Computing (CIT/IUCC/DASC/PICOM), Liverpool, UK, 26–28 October 2015; pp. 1596–1599.
Rathod, N.; Jain, P.; Subramanian, R.; Yawalkar, S.; Sunkenapally, M.; Amrutur, B.; Sundaresan, R.
Performance analysis of wireless devices for a campus-wide iot network. In Proceedings of the 2015
13th International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks
(WiOpt), Mumbai, India, 25–29 May 2015; pp. 84–89.
Centenaro, M.; Vangelista, L.; Zanella, A.; Zorzi, M. Long-Range Communications in Unlicensed Bands:
The Rising Stars in the IoT and Smart City Scenarios, 2015. Available online:
(accessed on 2 August 2016).
Petajajarvi, J.; Mikhaylov, K.; Roivainen, A.; Hanninen, T.; Pettissalo, M. On the coverage of lpwans:
Range evaluation and channel attenuation model for lora technology. In Proceedings of the 2015 14th
International Conference on ITS Telecommunications (ITST), Copenhagen, Denmark, 2–4 December 2015;
pp. 55–59.
Bor, M.; Vidler, J.E.; Roedig, U. Lora for the Internet of Things. In Proceedings of the 2016
International Conference on Embedded Wireless Systems and Networks (EWSN ’16), TU Graz, Austria,
15–17 February 2016; pp. 361–366
17. Berni, A.J.; Gregg, W.D. On the utility of chirp modulation for digital signaling. IEEE Trans. Commun. 1973,
21, 748–751.
Sensors 2016,16, 1466 18 of 18
Springer, A.; Gugler, W.; Huemer, M.; Reind, L.; Ruppel, C.; Weigel, R. Spread spectrum communications
using chirp signals. In Proceedings of the IEEE/AFCEA Information Systems for Enhanced Public Safety
and Security (EUROCOMM 2000), Munich, Germany, 19 May 2000; pp. 166–170.
LoRa SX1276/77/78/79 Datasheet, Rev. 4. Semtech, 2015. Available online:
images/datasheet/sx1276_77_78_79.pdf (accessed on 8 September 2016).
LoRa SX1272/73 Datasheet. Semtech, March 2015. Available online:
datasheet/sx1272.pdf (accessed on 8 September 2016).
Sikken, B. Project DecodingLoRa, 2016. Available online: (accessed on
1 August 2016).
International Telecommunication Union. P-Series Recommendations Radiowave Progapatin. Available online: (accessed on 1 August 2016).
LoRaWAN Specification V1.0. LoRa Alliance, 2015. Available online:
portals/0/specs/LoRaWAN%20Specification%201R0.pdf (accessed on 1 August 2016).
24. Abramson, N. The Aloha System: Final Technical Report; Hawaii University: Honolulu, HI, USA, 1974.
Oliver, M.; Escudero, A. Study of Different CSMA/CA IEEE 802.11-Based Implementations. Available online: (accessed on 8 September 2016).
2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (
... The long range (LoRa), as the name implies, is a proprietary long-range wide area wireless communication technology with a coverage range of up to 3 km in suburban areas with dense residential dwellings, as evaluated in [93]. LoRa operates in the unlicensed ISM band and is a physical layer protocol, suitable for low power, low throughput, and long-range network applications, like the IoT. ...
... Similar to LoRa, SigFox is suitable for the IoT, due to its long battery life, low device cost, high network capacity, and long range. The low power consumption is achieved by utilizing the duty cycles; thus, a wireless node wakes up and transmits, then waits, listens for a short duration, and returns to sleep mode [93]. ...
... ZigBee operates in the unlicensed ISM band at 2.4 GHz, offers data rates up to 250 kbits [93], and supports simple devices with minimal power operating in the personal operating space (POS) of 10m. It also provides self-organized, multi-hop and reliable mesh networking with long battery lifetime. ...
Full-text available
In recent years, tremendous advances have been made in the design and applications of wireless networks and embedded sensors. The combination of sophisticated sensors with wireless communication has introduced new applications, which can simplify humans’ daily activities, increase independence, and improve quality of life. Although numerous positioning techniques and wireless technologies have been introduced over the last few decades, there is still a need for improvements, in terms of efficiency, accuracy, and performance for the various applications. Localization importance increased even more recently, due to the coronavirus pandemic, which made people spend more time indoors. Improvements can be achieved by integrating sensor fusion and combining various wireless technologies for taking advantage of their individual strengths. Integrated sensing is also envisaged in the coming technologies, such as 6G. The primary aim of this review article is to discuss and evaluate the different wireless positioning techniques and technologies available for both indoor and outdoor localization. This, in combination with the analysis of the various discussed methods, including active and passive positioning, SLAM, PDR, integrated sensing, and sensor fusion, will pave the way for designing the future wireless positioning systems.
... To this end, extra focus is given to the Long Range (LoRa)/LoRa Wide Area Network (LoRaWAN) systems, which appear to be a prominent solution for wireless access over unlicensed spectrum. Taking into account the existing surveys in the area ( [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27]), we strive for an up-to-date comprehensive review on LPWANs, by providing valuable insights on the underlying technologies and state-of-the-art directions for research and development. Fig.1 reflects the adopted approach; with the solid-colored boxes, a chain of core sections is depicted, defining a main reading path for a reader familiar with the topic. ...
... Such distances are achieved thanks to robust Physical layer (PHY) protocols. More specifically, the 10 3 m is the main LPWAN operational range and refers to deployments in urban/suburban environments; the 10 4 m is an achievable range in areas that are mostly obstacle-free, Line of Sight (LOS) links, and rural environments [11]- [27]. Also, the 10 5 m range has been shown feasible in experimental setups. ...
... • Network performance validation and evaluation metrics (e.g., [11]) • Range evaluation and link-level behavior (e.g., [13]) • Power consumption, energy management, and lifetime of the network (e.g., [81]) • Multiple Access (e.g., [99]) • ADR algorithm (e.g., [122]) • ED mobility support and roaming (e.g., [123]) • Security in PHY, MAC, and Network layer (NET) (e.g., [130]) • Optimal GW placement and localization challenges (e.g., [141]) • Data-based decision making, by incorporating ML (e.g., [142]) • Meeting recent advancements from Information Technology (IT) and Telco domains, e.g., Software-Defined Networking (SDN), Blockchain, and network slicing (e.g., [143]) The research work on addressing challenges related to the above-listed topics includes approaches for all the wellknown steps in research methodology [19], [144]; namely: mathematical analysis, simulation study, and real-life system implementation. ...
This publication offers a comprehensive survey on the topic of Low-Power Wide Area Networks (LPWANs). We discuss the scope of LPWAN connectivity and present numerous well-established technologies that can offer such a service. From this presentation, LoRa/LoRaWAN systems emerge as one of the dominant solutions, and hence we offer details about end-to-end LoRa/LoRaWAN-enabled networking. As an area being in its infancy, LPWANs in general and LoRa/LoRaWAN in specific undergo extensive research. We shed light in key contributions, whereas we also provide our findings regarding the main research focus and needs. Simulation studies appear to be a popular choice among researchers, whereas real system implementations are constantly emerging thanks to low deployment and operational cost. For the former, we offer a comparative overview of existing simulators, whereas for the latter we discuss key considerations about the server-side part of a deployed system, which is usually supported by cloud infrastructure. Finally, LPWANs are systems that experience intense data generation, and thus we deliver a presentation of data-related operations which take place in such systems. This presentation revealed a major vertical market that is expected to be served by LPWANs: the Industiral IoT; which embodies concepts like Asset Administration Shell and Digital Twins.
... As shown in Figure 2, a LoRa packet is usually comprised of three parts: preamble, start frequency delimiter (SFD), and payload. The preamble contains a variable preamble of 6∼65535 baseline up-chirp to determine the start of the LoRa packet and a sync word of 2 up-chirps to differentiate the LoRa network [16]. The SFD contains 2.25 baseline down-chirps to indicate the start of the payload. ...
... Many methods have been proposed to analyze the whole process of LoRa demodulation. Augustin et al. [16] provides an overview of LoRa and an in-depth analysis of its functional components. Kang et al. [17] thoroughly analyzes the performance of the preamble detection model in the LoRa demodulation process and develops a optimal preamble detection scheme to maximum the detection probability. ...
Full-text available
LoRa has been shown as a promising Low-Power Wide Area Network (LPWAN) technology to connect millions of devices for the Internet of Things by providing long-distance low-power communication when the SNR is very low. Real LoRa networks, however, suffer from severe packet collisions. Existing collision resolution approaches introduce a high SNR loss, i.e., require a much higher SNR than LoRa. To push the limit of LoRa collision decoding, we present AlignTrack, the first LoRa collision decoding approach that can work in the SNR limit of the original LoRa. Our key finding is that a LoRa chirp aligned with a decoding window should lead to the highest peak in the frequency domain and thus has the least SNR loss. By aligning a moving window with different packets, we separate packets by identifying the aligned chirp in each window. We theoretically prove this leads to the minimal SNR loss. In practical implementation, we address two key challenges: (1) accurately detecting the start of each packet, and (2) separating collided packets in each window in the presence of CFO and inter-packet interference. We implement AlignTrack on HackRF One and compare its performance with the state-of-the-arts. The evaluation results show that AlignTrack improves network throughput by 1.68 $\times$ compared with NScale and 3 $\times$ compared with CoLoRa.
... LoRaWAN is a network protocol designed for applications that send limited quantities of data across great distances only a few times per 24 hours. Its low-power technology allows it to run for up to ten years without recharging [13]. ...
... As shown in the figure, all measurements show a distribution very close to the theoretical distribution of Rayleigh fading, except for the extremely low RSSI region of less than −120 dBm in Measurement #4, which is close to the RF sensitivity of LoRa chip SX1276 [33]. In [35], the minimum observed RSSIs were reported as higher than the specified RF sensitivity. Therefore, in these regions, the received RSSIs measured by the LoRa chip were considered relatively inaccurate. ...
Full-text available
Indoor private (local) wireless networks are considered key enablers for the industrial Internet of Things (IIoT). To ensure that a private network satisfies its own requirements and guarantees coverage, it is important to accurately measure the radio performance at the accurate position to further optimize the network. Additionally, as carrier frequencies increase, cellular network-based positioning becomes increasingly important, as it enables more accurate positioning. For example, centimeter-level accuracy is expected in beyond 5G systems owing to the sub-THz band carrier frequency. To accurately compare and assess various cellular network-based positioning methods, positioning devices, which can move along the same trajectory more accurately than at the centimeter-level, are required. To this end, we implemented a three-dimensional (3D) positioning device developed from open-source hardware, which is generally used to implement computer numerical control (CNC) machines and 3D printers. The implemented 3D pointing device is cost-effective and portable and can be freely configurable and resized in one, two, and three dimensions owing to the use of open-source hardware. The received signal strength indicator (RSSI) employing the LoRa module can be obtained using the implemented device. Using a very low carrier frequency system, it is possible to accurately measure the RSSI variations due to multipath propagation. After demonstrating the reproducibility of the data, the effectiveness of the implemented device was confirmed by showing that the RSSI variation and fading correlation coefficient are close to the Rayleigh fading variation.
Full-text available
The Internet of Things (IoT) concept involves connecting devices to the internet and forming a network of objects that can collect information from the environment without human intervention. Although the IoT concept offers some advantages, it also has some issues that are associated with cyber security risks, such as the lack of detection of malicious wireless sensor network (WSN) nodes, lack of fault tolerance, weak authorization, and authentication of nodes, and the insecure management of received data from IoT devices. Considering the cybersecurity issues of IoT devices, there is an urgent need of finding new solutions that can increase the security level of WSNs. One issue that needs attention is the secure management and data storage for IoT devices. Most of the current solutions are based on systems that operate in a centralized manner, ecosystems that are easy to tamper with and provide no records regarding the traceability of the data collected from the sensors. In this paper, we propose an architecture based on blockchain technology for securing and managing data collected from IoT devices. By implementing blockchain technology, we provide a distributed data storage architecture, thus eliminating the need for a centralized network topology using blockchain advantages such as immutability, decentralization, distributivity, enhanced security, transparency, instant traceability, and increased efficiency through automation. From the obtained results, the proposed architecture ensures a high level of performance and can be used as a scalable, massive data storage solution for IoT devices using blockchain technologies. New WSN communication protocols can be easily enrolled in our data storage blockchain architecture without the need for retrofitting, as our system does not depend on any specific communication protocol and can be applied to any IoT application.
Full-text available
Nowadays, it has become a crucial task in transferring confidential data for military departments, many multinational companies, etc. The important requirement is that the data that has been transmitted should not be visible to hackers or third parties from another end. To satisfy this requirement a wireless technology LoRa is used. Long-distance and low-power wireless communication technologies such as LoRa, Sigfox, and Narrowband-Internet of Things (NB-IoT) were developed in recent years. These technologies can contribute to indoor and outdoor smart applications with minimal power consumption. In this study, the LoRa wireless communication technique was used as the primary data communication method, enabling the device to communicate without requiring an Internet connection or a SIM card. This technology can be implemented in military and defense areas.
Full-text available
New transceiver technologies have emerged which enable power efficient communication over very long distances. Examples of such Low-Power Wide-Area Network (LPWAN) technologies are LoRa, Sigfox and Weightless. A typical application scenario for these technologies is city wide meter reading collection where devices send readings at very low frequency over a long distance to a data concentrator (one-hop networks). We argue that these transceivers are potentially very useful to construct more generic Internet of Things (IoT) networks incorporating multi-hop bi-directional communication enabling sensing and actuation. Furthermore, these transceivers have interesting features not available with more traditional transceivers used for IoT networks which enable construction of novel protocol elements. In this paper we present a performance and capability analysis of a currently available LoRa transceiver. We describe its features and then demonstrate how such transceiver can be put to use efficiently in a wide-area application scenario. In particular we demonstrate how unique features such as concurrent non-destructive transmissions and carrier detection can be employed. Our deployment experiment demonstrates that 6 LoRa nodes can form a network covering 1.5ha in a built up environment, achieving a potential lifetime of 2 year on 2 AA batteries and delivering data within 5s and reliability of 80%.
Conference Paper
Full-text available
In addition to long battery life and low cost, coverage is one of the most critical performance metrics for the low power wide area networks (LPWAN). In this work we study the coverage of the recently developed LoRa LPWAN technology via real-life measurements. The experiments were conducted in the city of Oulu, Finland, using the commercially available equipment. The measurements were executed for the cases when a node located on ground (attached on the roof rack of a car) or on water (attached to the radio mast of a boat) was reporting its data to a base station. For a node operating in the 868 MHz ISM band using 14 dBm transmit power and the maximum spreading factor, we have obtained the maximum communication range of over 15 km on ground and close to 30 km on water. Besides the actual measurements, in the paper we also present a channel attenuation model derived from the measurement data. The model can be used to estimate the path loss in 868 MHz ISM band in an area similar to Oulu, Finland.
Full-text available
Connectivity is probably the most basic building block of the Internet of Things (IoT) paradigm. Up to know, the two main approaches to provide data access to the \emph{things} have been based either on multi-hop mesh networks using short-range communication technologies in the unlicensed spectrum, or on long-range, legacy cellular technologies, mainly 2G/GSM, operating in the corresponding licensed frequency bands. Recently, these reference models have been challenged by a new type of wireless connectivity, characterized by low-rate, long-range transmission technologies in the unlicensed sub-GHz frequency bands, used to realize access networks with star topology which are referred to a \emph{Low-Power Wide Area Networks} (LPWANs). In this paper, we introduce this new approach to provide connectivity in the IoT scenario, discussing its advantages over the established paradigms in terms of efficiency, effectiveness, and architectural design, in particular for the typical Smart Cities applications.
Full-text available
Available online: Smart technologies play a key role in sustainable economic growth. They transform houses, offices, factories, and even cities into autonomic, self-controlled systems acting often without human intervention and thus sparing people routine connected with information collecting and processing. The paper gives an overview of a novel Wi-Fi technology, currently under development, which aims to organize communication between various devices used in such applications as smart grids, smart meters, smart houses, smart healthcare systems, smart industry, etc.
Full-text available
M2M communications are projected to be one of the fastest growing technology segments of the IT sector in the next years. Sensor and actuator networks connect communication machines and devices so that they automatically transmit information, serving the growing demand for environmental data acquisition. IEEE 802.11ah Task Group addresses the creation of a new standard for giving response to the particular requirements of this type of networks: large number of power-constrained stations, long transmission range, small and infrequent data messages, low data-rates and non-critical delay. This article explores the key features of this new standard under development, especially those related to the reduction of energy consumption in the MAC Layer. In this direction, a performance assessment of IEEE 802.11ah in four typical M2M scenarios has been performed.
Full-text available
433MHz is getting more attention for Machine-to-Machine communication. This paper presents the DASH7 Alliance Protocol, an active RFID alliance standard for 433MHz wireless sensor communication based on the ISO/IEC 18000-7. First, the major differences of 433MHz communication compared to more frequently used frequencies, such as 2.4GHz and 868/920MHz are explained. Subsequently, the general concepts of DASH7 Alliance Protocol are described, such as the BLAST networking topology and the different OSI layer implementations, in a top-down method. Basic DASH7 features such as the advertising protocol, ad-hoc synchronization and query based addressing are used to explain the different layers. Finally, the paper introduces a software stack implementation named OSS-7, which is an open source implementation of the DASH7 alliance protocol used for testing, rapid prototyping, and demonstrations.