LoRa for the Internet of Things

Martin Bor John Vidler Utz Roedig

Department of Computing and Department of Computing and Department of Computing and
Communications Communications Communications
Lancaster University Lancaster University Lancaster University
[email protected] [email protected] [email protected]

Abstract distances. Examples of such LPWAN technologies are

New transceiver technologies have emerged which enable LoRa [1], Sigfox [2] and Weightless [3]. These new
power efficient communication over very long distances. Ex- transceiver types target applications where thousands of de-
amples of such Low-Power Wide-Area Network (LPWAN) vices are used in a large geographic area to collect sensor
technologies are LoRa, Sigfox and Weightless. A typical readings. A typical application is the collection of meter
application scenario for these technologies is city wide me- readings in a city. These systems are used in a setup where
ter reading collection where devices send readings at very simple devices send data in one hop to powerful receiver
low frequency over a long distance to a data concentra- which then forwards data over a fixed wired infrastructure
tor (one-hop networks). We argue that these transceivers to a data collection point.
are potentially very useful to construct more generic Inter- We argue that these transceivers are potentially very use-
net of Things (IoT) networks incorporating multi-hop bi- ful to construct more generic IoT networks incorporating
directional communication enabling sensing and actuation. multi-hop bi-directional communications enabling sensing
Furthermore, these transceivers have interesting features not and actuation. The transceivers have the ability to com-
available with more traditional transceivers used for IoT net- municate over large distances on a small energy budget
works which enable construction of novel protocol elements. which would enable us to build more efficient IoT infras-
In this paper we present a performance and capability tructures than currently possible. For example, commonly
analysis of a currently available LoRa transceiver. We de- used ZigBee transceivers such as the Chipcon CC2420 cover
scribe its features and then demonstrate how such transceiver a communication range of 20 m using 84.5 µJ (40 byte mes-
can be put to use efficiently in a wide-area application sce- sage) in a built up environment. A LoRa Semtech SX1272
nario. In particular we demonstrate how unique features transceiver can cover a distance of 150 m using 86.5 mJ in
such as concurrent non-destructive transmissions and car- the same environment.
rier detection can be employed. Our deployment experiment Besides improved communication range, the transceivers
demonstrates that 6 LoRa nodes can form a network covering have unique features stemming from the used modulation
1.5 ha in a built up environment, achieving a potential life- schemes. Thus, it is not efficient to simply use these
time of 2 year on 2 AA batteries and delivering data within transceivers with existing Medium Access Control (MAC)
5 s and reliability of 80%. protocols and routing mechanisms that have emerged in
Categories and Subject Descriptors the IoT domain. When construction a network using these
C.2.1 [Network Architecture and Design]: Wireless transceivers their specific capabilities should be taken into
communication account to maximise performance in terms of communica-
Keywords tion and minimise energy consumption.
LoRa, IoT, Medium Access Control (MAC) In this paper we investigate LoRa as technology for build-
ing generic IoT networks. We investigate the communication
1 Introduction capability of the LoRa Semtech SX1272 transceiver and its
Recently new transceiver technologies have emerged energy consumption patterns. We analyse in detail unique
which enable power efficient communication over very long communication features offered by the transceiver. A de-
tailed analysis of these features is essential as they should
be exploited when constructing communication protocols on
top of this communication technology. Finally we construct
an example communication protocol for the LoRa physical
layer which enables wide-area multi-hop data collection and
actuation without existing backbone infrastructure. Our de-
ployment experiment demonstrates that 6 LoRa nodes can
form a network covering 1.5 ha in a built up environment,
achieving a potential lifetime of 2 years on 2 AA batteries
and delivering data within 5 s and reliability of 80%.
 The specific contributions of this paper are: 2.2 LoRa Physical Layer
• LoRa Feature Evaluation: We describe and analyse LoRa is a physical layer specification based on CSS with
LoRa specific physical layer capabilities. This includes integrated Forward Error Correction (FEC). Transmissions
channel separation using different Spreading Factor use a wide band to counter interference and to handle fre-
(SF), behaviour regarding non-destructive concurrent quency offsets due to low cost crystals. A LoRa receiver can
transmission; Clear Channel Assesment (CCA) in form decode transmissions 19.5 dB below the noise floor. Thus,
of Carrier Activity Detection (CAD). very long communication distances can be bridged. LoRa
key properties are: long range, high robustness, multipath
• LoRaBlink: We describe a protocol (LoRaBlink) on resistance, Doppler resistance, low power. LoRa operates
top of the LoRa’s physical layer which exploits LoRa’s in the lower ISM bands (EU: 868 MHz and 433 MHz, USA:
unique features. The protocol enables energy efficient 915 MHz and 433 MHz).
wide-area multi-hop data collection. A LoRa radio has four configuration parameters: carrier
• LoRaBlink Performance Evaluation: We provide an frequency, spreading factor, bandwidth and coding rate. The
evaluation of LoRaBlink in a small testbed using selection of these parameters determines energy consump-
6 nodes equipped with a Semtech SX1272 transceiver. tion, transmission range and resilience to noise. In the fol-
lowing sections we use the Semtech SX1272 transceiver as
The next section describes our evaluation platform. Sec-
reference point.
tion 2 gives an overview of the LoRa Physical and MAC
layer and Section 3 describes our LoRa evaluation platform. Carrier Frequency
Section 4 gives an evaluation of LoRa specific features and Carrier Frequency (CF) is the centre frequency used for
Section 5 describes our proposed protocol LoRaBlink. Sec- the transmission band. For the SX1272 it is in the range
tion 6 provides the evaluation of LoRaBlink and Section 7 of 860 MHz to 1020 MHz, programmable in steps of 61 Hz.
concludes the paper. We do not provide a dedicated related The alternative radio chip Semtech SX1276 allows adjust-
work section as so far the research community has not in- ment from 137 MHz to 1020 MHz.
vestigated LoRa in depth and to the best of our knowledge
Spreading Factor
no other research papers exist describing LoRa’s capabilities
SF is the ratio between the symbol rate and chip rate.
and deployments.
A higher spreading factor increases the Signal to Noise Ra-
tio (SNR), and thus sensitivity and range, but also increases
2 LoRa the air time of the packet. The number of chips per symbol is
LoRa (Long Range) is a proprietary spread spectrum calculated as 2sf . For example, with an SF of 12 (SF12) 4096
modulation technique by Semtech. It is a derivative of Chirp chips/symbol are used. Each increase in SF halves the trans-
Spread Spectrum (CSS). The LoRa physical layer may be mission rate and, hence, doubles transmission duration and
used with any MAC layer; however, LoRaWAN is the cur- ultimately energy consumption. Spreading factor can be se-
rently proposed MAC which operates a network in a simple lected from 6 to 12. SF6, with the highest rate transmission,
star topology. is a special case and requires special operations. For exam-
ple, implicit headers are required. Radio communications
2.1 LoRaWAN with different SF are orthogonal to each other and network
As LoRa is capable to transmit over very long distances separation using different SF is possible.
it was decided that LoRaWAN only needs to support a star
topology. Nodes transmit directly to a gateway which is pow- Bandwidth
ered and connected to a backbone infrastructure. Gateways Bandwidth (BW) is the range of frequencies in the trans-
are powerful devices with powerful radios capable to receive mission band. Higher BW gives a higher data rate (thus
and decode multiple concurrent transmissions (up to 50). shorter time on air), but a lower sensitivity (due to integration
Three classes of node devices are defined: (1) Class A end- of additional noise). A lower BW gives a higher sensitivity,
devices: The node transmits to the gateway when needed. but a lower data rate. Lower BW also requires more accurate
After transmission the node opens a receive window to ob- crystals (less ppm). Data is send out at a chip rate equal to
tain queued messages from the gateway. (2) Class B end- the bandwidth. So, a bandwidth of 125 kHz corresponds to a
devices with scheduled receive slots: The node behaves like chip rate of 125 kcps. The SX1272 has three programmable
a Class A node with additional receive windows at sched- bandwidth settings: 500 kHz, 250 kHz and 125 kHz. The
uled times. Gateway beacons are used for time synchroni- Semtech SX1272 can be programmed in the range of 7.8 kHz
sation of end-devices. (3) Class C end-devices with maxi- to 500 kHz, though bandwidths lower than 62.5 kHz requires
mal receive slots: these nodes are continuous listening which a temperature compensated crystal oscillator (TCXO).
makes them unsuitable for battery powered operations. Coding Rate
In this paper we propose an alternative MAC for LoRa Coding Rate (CR) is the FEC rate used by the LoRa mo-
(LoRaBlink as described in Section 5) which enables multi- dem and offers protection against bursts of interference. A
hop communication in a network of battery operated and higher CR offers more protection, but increases time on air.
duty-cycled devices. Although star networks with a powered Radios with different CR (and same CF/SF/BW), can still
and powerful gateway device are an option in some situations communicate with each other. CR of the payload is stored in
it does not cover all IoT application scenarios. the header of the packet, which is always encoded at 4/8.
 If we assume a system where the above message is trans-
mitted every 15 min and we assume a battery capacity of 2
typical AA batteries of 5400 mA h the node will have a life-
time of 6.2 years.
If we assume a node only carries out a CAD every 5 s to
check for an incoming message the node will have a lifetime
of 6.0 years (assuming again 5400 mA h battery capacity).
4 LoRa Feature Evaluation
LoRa has interesting features aside the increased com-
munication range that should be taken into account when
Figure 1. NetBlocks XRange SX1272 LoRa RF module. constructing network protocols. For example, channel
separation using different SF is possible, concurrent non-
destructive transmissions are possible and carrier detection
2.3 LoRa Characteristics via CAD is provided. However, from available documenta-
Using a LoRa radio in a sensor network has some inter- tion the performance and ability of these features is not clear.
esting aspects. First, since the range is relatively large (hun- Therefore we carry out a series of experiments to evaluate
dreds of meter indoors, kilometres outdoors), networks can these provided features.
span large areas without routing over many hops. In many 4.1 Spreading Factors
cases one hop from every node to the sink is feasible. Sec- Different spreading factors are claimed to be orthogonal
ondly, transmission on the same carrier frequency, but with to each other. Thus, construction of virtual channels on the
different spreading factor, are orthogonal. This creates the same carrier frequency is possible (Code Division Multiple
opportunity of dividing the channel in virtual subchannels. Access (CDMA)).
Thirdly, when transmissions occur at the same time with the We evaluate how well this separation works using a sim-
same parameters, the strongest transmission will be received ple experiment setup. A transmitter is set to continuously
with high probability, ie. concurrent transmissions are non- send a 40 B packet with a fixed SF. A receiver set to the same
destructive even when their contents is different. This feature SF is used to receive the transmissions. A second transmitter
is exploited by LoRaWAN where all gateways broadcast bea- is used to transmit continuously and sequentially using all
cons at the same time (tight clock synchronisation via GPS) other SF to the same receiver.
and an end device is able to demodulate the strongest bea-
con.
Findings
All transmissions where sender and receiver use the same
3 LoRa Experimental Device SF are received correctly. None of the transmissions emitted
For our studies we use the XRange device from Net- by the second node using a different SF are received. This
Blocks1 as shown in Figure 1. The device comprises a result suggests that channel separation using SF works per-
SX1272 LoRa transceiver and a low-power STM32L151 fectly.
ARM micro-controller. We use GCC ARM and the LoRa However, as we will show in Section 4.3 this is only par-
radio driver and runtime environment derived from the IBM tially true. When using CAD to detect an incoming transmis-
LMiC (LoRaWAN in C)2 . The runtime environment, code sion a signal using the wrong SF may be detected as valid
used for our experiments and the communication proto- transmission even though it cannot be decoded. This is im-
col presented in Section 5 are available at http://www. portant as the false detection rate has a negative impact on
lancaster.ac.uk/scc/sites/lora/. energy efficiency of a protocol.
The energy consumption of the system is for most appli- 4.2 Concurrent Transmissions
cation cases dominated by energy cost for communications. In LoRa concurrent transmissions are claimed to be non-
Energy consumption for transmission, reception, listening destructive and such feature is very valuable for protocol de-
and CAD must be distinguished. The energy consumption in sign. Well-timed cooperative transmissions have been used
these states depends on selected SF and BW. Also, selected in Glossy [6]. In Glossy the same message is transmitted
communication parameters will influence transmission times accurately timed by multiple nodes allowing correct recep-
of packets and ultimately energy consumption. tion. A-MAC [5] and also Whitehouse et al. [7] make use
To give an example we assume SF12, BW125, CR4/5, of the capture effect. Here multiple different messages are
and TX power 17 dBm (An energy hungry setting allowing transmitted concurrently and depending on power levels and
very long ranges which was used in our experimental evalu- timing one of the concurrently transmitted messages can be
ation discussed later). A transmission of a packet with 10 B received.
payload and 12.25 symbols preamble has a transmission du- We set up an experiment to understand the exact con-
ration of 991.23 ms. Transmitting such message will cost ditions in which this effect is present in LoRa. We use
214 mJ. Reception of this message will cost 25.7 mJ and per- a receiver, one ‘weak’ transmitter and one ‘strong’ trans-
forming a CAD will cost 1.23 mJ. mitter (1 dBm difference). Both transmitters send a packet
with explicit header and CRC. The strong transmitter varied
1 http://www.netblocks.eu/ the transmission time offset relative to the weak transmitter.
2 http://www.research.ibm.com/labs/zurich/ics/lrsc/lmic.html From being one packet (airtime) early to being one packet
 1.0 1
0.8
PRR

weak
0.6 500 0.8

Bandwidth (kHz)
0.4
0.2
0.0 0.6
1.0
0.8
250

strong
PRR

0.6 0.4
0.4
0.2
0.0 125 0.2
1.0
0.8 0

corrupt
PRR

0.6
0.4 7 8 9 10 11 12
0.2 Spreading Factor
0.0
-80 -60 -40 -20 0 20 40 60 80 Figure 3. Carrier detection ratios for a transmitter send-
Offset (Symbol time) ing at spreading factor 7 and bandwidth 250 kHz, indi-
Figure 2. Example collision result. Spreading factor 11, cated by the white cross. Carriers were also detected by
bandwidth 125 kHz. X-axis shows the transmission offset adjacent data rates.
relative to the weak node in symbol time, Y-axis shows
the packet reception rate.
Findings
One of two concurrent transmission can be received with
very high probability if both transmissions do not have an
(airtime) late. For each offset, 16 32-byte packets were trans-
offset of more than 3 symbol periods. This translates to a
mitted using first identical packet payloads and subsequently
duration between 768 µs and 98.3 ms, depending on the SF
different payloads. We also run the experiment with all com-
and BW. Synchronisation of nodes within these bounds is
binations of SF and BW.
relatively easy to achieve and therefore protocols making use
The experiment results are shown in Figure 2 for SF12 of this feature can easily be implemented with LoRa.
and BW 125 kHz. The Y-axis represents the packet reception
rate. The X-axis represents the transmission offset relative to 4.3 Carrier Activity Detection
the weak node in symbol time. The top bar shows packets re- Transceivers normally provide a CCA interface to detect
ceived from the weak transmitter at the receiver; the middle an occupied channel. The CCA is used in communication
bar shows packets received from the strong transmitter at the protocols to decide if packets can be transmitted and to de-
receiver. The bottom bar shows when packets were received cide if the radio must be kept active to receive a message. In
from either transmitter but deemed corrupt (Cyclic Redun- particular for power constrained nodes it is important to have
dancy Check (CRC) failure). We did notice that about 1 in an accurate and fast CCA mechanism as this enables imple-
6000 packets was corrupted, but did not fail the CRC. Often mentation of power-efficient duty cycling. Nodes perform
these packets had 1 bit corrupted. periodic short CCA checks and only power the receiver for
longer if a transmission is detected.
Results for other SF and BW combinations are very simi-
lar. Also, transmitting the same packet payload or a different LoRa transceivers do not provide a classical CCA inter-
payload does not change the obtained results significantly. face based on an Received Signal Strength Indicator (RSSI)
threshold to detect an occupied channel. LoRa can receive
As can be seen, the strong transmitter is successfully de- transmissions with a signal strength that is below the noise
coded if it transmits not later than 3 symbol periods after the floor and, consequently, an RSSI threshold check will not re-
weak transmitter started (symbol period is Tsym = 2SF /BW ). veal an occupied channel. LoRa radios provide therefore a
If the weak transmitter starts later than 3 symbol periods no CAD mode to detect a present preamble.
transmission is received (or corrupted data is received). The CAD process takes approximately 2 symbol periods,
Although the packet takes 60.25 symbol periods to trans- and only requires the radio on for about 1 symbol period
mit, both nodes can be received at an offset of −57 sym- (The exact CAD duration is calculated as sum of (32/BW +
bol periods or more. The tail of the strong node does destroy 2SF /BW) seconds in RX mode and (SF · 2SF )/(1750 × 103 )
the initial preamble of the weak node, but as long as at most 3 seconds of processing). The processing phase requires about
symbols are destroyed, the weak packet can also be success- half of the energy required in receive mode (around 6 mA de-
fully received. This relationship is not symmetrical, as at an pending on the SF/BW). After receiving a ‘CAD detected’,
offset of +57 symbol periods, the weak node’s tail (CRC) a transceiver can be switched in RX mode to receive the on-
gets destroyed, invalidating a packet which may have been going transmission if required.
correctly received. It is only that at an offset of +60 sym- We set up an experiment to test the reliability of CAD.
bol periods or more, both packets gets received perfectly. A detector node starts the CAD process on a regular inter-
We also experimented with two transmitters set to the val (every 100 ms) and records whether it has detected a
same transmit power. In this case either of the two is per- carrier or not. After 300 samples, it switches the SF/BW
ceived as stronger and the above described behaviour applies combination and repeats the process. A transmitter node is
(although the role of stronger/weaker transmitter may alter- programmed to continuously send out preambles at a fixed
nate with each transmission making it difficult to conduct SF/BW combination. The experiment is repeated with dif-
experiments and describe results). ferent transmitter SF/BW combinations.
 The results for a transmitter using SF7 and a BW of within slots and properties of the LoRa physical layer ensure
250 kHz are shown in Figure 3. Results for a transmitter us- that one of the concurrent transmissions is received. Mes-
ing different SF/BW combinations are similar. When trans- sages are distributed from the sink to nodes using flooding.
mitter and receiver use the same SF/BW combination the Messages from nodes to the sink use a directed flooding ap-
worst detection rate was measured at 97% (for SF7 and a proach. The result is a very simple but robust protocol which
BW of 250 kHz). However, the CAD process also detects covers the set requirements.
carriers in SF/BW combinations different from the combina- Figure 4 shows an operation example of LoRaBlink in a
tion the transmitter is using (up to 99% detections for SF9 network containing 3 nodes and a sink. Node 1 and 2 are in
and a BW of 500 kHz). This happens for data rates that are communication range of the sink. Node 3 cannot be directly
adjacent to the current data rate. When no transmitter is ac- reached by the sink but is in range of node 1 and 2.
tive, the detector had a false positive rate of 0.092%. Each node powering up will remain in listen mode until
Findings a beacon is received. Beacons are used for time synchroni-
CAD can only detect channel occupancy while a pream- sation and mark the start of an epoch. Each epoch contains
ble is transmitted. The detection probability is high (above N slots. The first NB slots of an epoch are used for beacon
97%) and false positives are low (0.092%). However, if mul- transmissions. A beacon message contains the hop distance
tiple LoRa networks are active on different SF/BW combina- to the sink and upon receiving a beacon a node will transmit
tions false positives can be very high (depending on SF/BW its own beacon according to its distance to the sink. A node
ratios). When using multiple SF/BW combinations in the will aim to select its position based on minimal distance to
same network (or when constructing multiple networks sep- the sink. In the example in Figure 4 the sink transmits a
arated by SF/BW) the choice of combinations is important beacon received by node 1 and node 2. Both nodes use the
when using CAD. beacon to determine epoch start and their distance to the sink
5 LoRaBlink (1 hop). In the next beacon slot node 1 and 2 transmit their
In this section we describe LoRaBlink, an IoT protocol beacon concurrently. Due to properties of the LoRa physical
for LoRa transceivers. The protocol is designed to support layer either one of these (depending on transmission time dif-
reliable and energy efficient multi-hop communication. It is ference and perceived signal strength at node 3) is received
also designed to support low-latency bidirectional commu- at node 3. Node 3 updates its hop distance to the sink as 2
nication. Thus, the described protocol is different to defined and transmits its own beacon in the next beacon slot. This
protocols for LoRa as given in the LoRaWAN 1.0 specifica- beacon is received by node 2 (we assume node 1 would not
tion. The protocol relies on features and building blocks as receive) which discards the message as its hop count is less
described in the previous section. than 2. The number of beacon slots NB determines the max-
imum depth the network can have.
5.1 Protocol Aims
LoRaBlink aims to address a number of aspects neces- Following the beacon slots are ND data slots. A node that
sary for deployment of IoT applications and which are not has data to transmit selects the next available data slot and
covered by currently defined LoRa protocols. These are: transmits. After transmission a node listens for an Acknowl-
edgement (ACK) (an optional protocol feature; the ACK is
• Multi-Hop: The protocol should support communica- not shown in Figure 4). Two nodes may transmit in the same
tion over multiple hops. slot with the result that at least one message is decoded by
• Low-Energy: Nodes should be able to duty-cycle to one receiver, with a chance that two different nodes in the
conserve energy and enable battery powered operations network decode one of each transmission. If a node has a
over long time spans. lower hop count to the sink than the source node it will relay
• Resilience: The protocol should be resilient and enable the message in the next slot. Multiple nodes may forward
high message delivery probability. which introduces redundancy. ACK messages may also col-
lide but a receiver will always be able to decode one of mul-
• Low-Latency: The protocol should enable low-latency tiple ACK correctly. In Figure 4 node 3 generates a data
communication. message. The message is received by node 2 and 1 which
Further to these requirements we also make the assump- then forward the message simultaneously in the next slot.
tion that the network has a low density, low traffic volume The sink will be able to decode one of the two transmissions.
and contains a limited number of nodes. We also assume Data travelling from the sink to a node will use the same
that a single sink is used for communication and that com- mechanism as used for beacon distribution. If the sink has
munication is between the sink and the nodes. to send non-delay sensitive information to nodes in the net-
A vast number of protocols exist to implement theses re- work it can be delayed and included in beacon messages for
quirements [4]. However, none of the available options is distribution.
particularly designed to make use of LoRa specific features
such as the ability to receive one message out of a pool of 5.3 Node Lifetime
concurrent transmissions. To improve energy consumption of the system beacon
5.2 Protocol Operations messages are sent infrequent (a long epoch is used) and the
The protocol integrates MAC and routing in a single sim- CAD is used within slots to detect incoming transmissions.
ple protocol. Time synchronisation among nodes is used to Infrequent beacon transmission is possible as tight time syn-
define slotted channel access. Nodes transmit concurrently chronisation in the network is not necessary.
 Sink TB RB C C C RD C C TB
TB Transmit Beacon
Node 1 Listen RB TB C C RD TD C C RB RB Receive Beacon
TD Transmit Data
Node 2 Listen RB TB RB C RD TD C C RB
RD Receive Data
Node 3 Listen RB TB C TD C C C C C CAD
Figure 4. LoRaBlink: Protocol example using a 4 node network.

power of 17 dBm. The epoch length was set to 5 min with

NB = 5 and ND = 55 (slots every 5 seconds). Nodes are set
to transmit a data packet (10 byte) randomly within one slot
of each epoch. In this experiment we did not use CAD and
instead implemented a listen period of 50 symbols in each
slot. This was done to avoid packet losses due to CAD and
to evaluate data delivery of LoRaBlink on its own.
In our evaluation packets from all nodes were delivered
with a reliability of 80% over a duration of 2.3 h. Node 4
delivered messages for the first half of the experiment from
the position marked 4a and later from position 4b. Node 3
and 4b delivered messages via one hop while all other nodes
were able to directly communicate with the sink node. Mes-
sages transmitted by node 3 and 4b are relayed by multiple
nodes (node 5, node 2 and node 4a for node 3) in the same
slot.
The experiment shows that LoRaBlink can deliver mes-
sages reliably over large distance in a challenging multi-hop
environment (structures in the communication path). The ex-
periment also shows that using concurrent transmissions is
feasible.
7 Conclusion
LoRa radios are capable of longer communication ranges
Figure 5. LoRaBlink: Map of a small scale deployment. than commonly used IoT radios while still being energy effi-
Lines are routes between nodes, with distance in meters. cient. In addition, these radios provide interesting features
such as non-destructive concurrent transmissions. As we
have shown, LoRa radios can be used in more general net-
Epoch length and NB and ND determine energy consump- work layouts than the one used by LoRaWAN. Thus, we be-
tion and data transport delay in the network. lieve that LoRa transceivers provide an interesting option for
While not the most energy efficient protocol, the addi- building general IoT applications.
tional range may be a benefit to low-node-density deploy- 8 References
ments, requiring far fewer forwarding nodes to cover the
[1] LoRa. https://www.lora-alliance.org. Accessed: 2015-11-07.
same area. The transceiver settings described in Section 3 [2] Sigfox. http://www.sigfox.com. Accessed: 2015-11-07.
and an epoch length of 15 min with NB = 3 and ND = 177 (5 [3] Weightless open standard. http://www.weightless.org. Accessed:
second slot) we obtain a maximum node lifetime of 2 years 2015-11-07.
with two AA batteries (5400 mA h) – assuming that one bea- [4] A. Bachir, M. Dohler, T. Watteyne, and K. K. Leung. Mac essentials for
con is transmitted and two are received and all other slots in wireless sensor networks. Communications Surveys & Tutorials, IEEE,
12(2):222–248, 2010.
the epoch contain one CAD. [5] P. Dutta, S. Dawson-Haggerty, Y. Chen, C.-J. M. Liang, and A. Terzis.
6 LoRaBlink Evaluation Design and evaluation of a versatile and efficient receiver-initiated link
We deployed a 6 node network on the university campus layer for low-power wireless. In Proceedings of the 8th ACM Confer-
ence on Embedded Networked Sensor Systems, SenSys ’10, pages 1–14,
as shown in Figure 5. Nodes are deployed in buildings across New York, NY, USA, 2010. ACM.
campus on the ground floor within buildings approximately [6] F. Ferrari, M. Zimmerling, L. Thiele, and O. Saukh. Efficient network
1.5 m above the floor. The sink node is located in the third flooding and time synchronization with glossy. In Information Process-
floor on a windowsill. Node 4 was first deployed at position ing in Sensor Networks (IPSN), 2011 10th International Conference on,
pages 73–84, April 2011.
4a and was moved to position 4b to create a larger network.
[7] K. Whitehouse, A. Woo, F. Jiang, J. Polastre, and D. Culler. Exploiting
Nodes have to communicate through several buildings and the capture effect for collision detection and recovery. In Embedded
structures. Networked Sensors, 2005. EmNetS-II. The Second IEEE Workshop on,
We use in the experiment SF12 and BW125 and a TX pages 45–52, May 2005.