Simulation-based Study of Interference Impact in ISM Bands in Smart

Cities: Connected Trafﬁc Light for Visually Impaired People Use-case

Mohammad Rahal

, Marc Ibrahim

and Gerard Chalhoub

Ecole Sup

erieure d’Ing

enieurs de Beyrouth, Saint Joseph University of Beirut, Lebanon

University Clermont Auvergne, LIMOS-CNRS, Clermont-Ferrand, France

Keywords:

Interference, ISM, Smart City, Simulation, Evaluation.

Abstract:

Wireless technologies operating in the unlicensed ISM (Industrial Scientiﬁc and Medical) bands are om-

nipresent. Indeed, with the expansion of the Internet of Things, plenty of applications are being developed

on devices that use the 868.3 - 868.8 MHz ISM band. This expansion has an impact on existing technologies

operating at the same frequencies. This paper focuses on the impact of IoT networks on a special use case

deployed in smart cities which is the connected trafﬁc light for visually impaired pedestrians. A connected

trafﬁc light is equipped with a radio receiver that operates at 868.3MHz frequency. It allows pedestrians to

query the state of the trafﬁc light using a handheld remote control. When the latter is pressed, a generated radio

message will activate a sound beacon that tells the state of the trafﬁc light. In some cases, when the trafﬁc light

is not optimally deployed or when it suffers from interference, the connection with the remote control cannot

be established. In this paper, we investigate, using a detailed simulation analysis, the impact of interfering IoT

devices and antenna types used on the receiver module of the trafﬁc light on the quality of the radio links.

1 INTRODUCTION

The past decade has witnessed a large growth in

the use of Low-Power Wide-Area (LPWA) technolo-

gies, which are low cost solutions providing long-

range wireless communications. Long range sub-

GHz band technologies such as LoRa, SigFox and

IEEE 802.15.4 are getting increasingly popular for

academic research and daily life applications (Saelens

et al., 2019) (Augustin et al., 2016). This growth has

a considerable impact on existing applications using

the same unlicensed Industrial Scientiﬁc and Medical

(ISM) bands. One such application is connected traf-

ﬁc lights for visually impaired people.

The use of ISM frequency bands is tightly regu-

lated in European Union (EU) and it must conﬁrm

to Short Range Devices (SRDs) regulations. Even

though these regulations are respected, sharing of the

spectrum between technologies operating in free ISM

Band raises serious problems, leading to a degrada-

tion of performance, dropping of packets, decreasing

of the throughput and malfunctioning of the whole

network. Interference could affect the security and

the functionality of the network. Due to these reasons

mentioned before, transmission of the modulated sig-

nal will be subjected to interference from the grow-

ing number of devices which can degrade Signal-To-

Noise and Interference Ratio (SINR), leading to a

greater Bit-Error-Rate BER or in most cases to packet

loss.

NF-S32002/A1 protocol (Minaudier et al., 2006)

is one of the technologies that comply to the SRDs

ETSI standard since it operates at the band centered at

868.3MHz. This protocol is used to help visually im-

paired people to communicate with connected pedes-

trian trafﬁc lights in order to know its current state.

Indeed, a visually impaired person will send a radio

signal to the trafﬁc light using a remote control. The

trafﬁc light will reply with an sound signal indicating

its current state. This will help the person to know

when it is safe to cross the street.

In this paper, we investigate the impact of ISM

interference from technologies using the same fre-

quency band as NF-S32002/A1, such as LoRa and

Sigfox, on the trafﬁc light receivers that are deployed

in Clermont-Ferrand city center. Consequently, we

use simulations to quantify the loss of the signals sent

by pedestrians depending on the SINR as well as the

bandwidth of the ISM interfering signal. We also

study the improvements that the use of directional

antennas on the trafﬁc light receiver could bring to

the radio link. We demonstrate that directional anten-

nas increasing the resiliency of these receivers against

Rahal, M., Ibrahim, M. and Chalhoub, G.

Simulation-based Study of Interference Impact in ISM Bands in Smart Cities: Connected Trafﬁc Light for Visually Impaired People Use-case.

DOI: 10.5220/0010575300250032

In Proceedings of the 18th International Conference on Wireless Networks and Mobile Systems (WINSYS 2021), pages 25-32

ISBN: 978-989-758-529-6

ISM interference. We thus show that as the SINR

situation improves by mitigating harmful interference

the BER performance will be better. We also specify

the threshold levels of SINR that should be respected

to have low or zero packet loss. We do not propose

any modiﬁcations to the existing communication stan-

dards, we focus on the added value of using a simu-

lation study in order to give deployment guidelines of

these wireless systems.

The remainder of this paper is organized as fol-

lows. Section II illustrates a brief overview of pre-

vious research about ISM interference of co-existing

technologies. Section III presents the NF-S32002/A1

protocol. In Section IV we present HTZ simulator and

describe the scenarios we used to assess the impact of

ISM interference on connected trafﬁc light. Simula-

tion results are then presented in Section V. Finally,

Section VI is dedicated for conclusions and perspec-

tives. d.

2 STATE OF THE ART

Modulated signals will suffer from co-technology in-

terference as well as inter-technology interference

when operating in ISM bands. Inter-technology ISM

interference is attributed to the co-existence of mul-

tiple technologies that use the same frequencies of

the unlicensed shared ISM bands. To the best of our

knowledge, this is the ﬁrst paper that studies the in-

terference between different long range wireless IoT

systems and a short range wireless technology used

by pedestrian trafﬁc light receivers.

In (Lauridsen et al., 2017), authors in focused on

determining if there is any interference which may

impact deployment of Internet of things. Their focus

was on LoRa and Sigfox technologies. Another arti-

cle (Elshabrawy and Robert, 2018) studied the evalu-

ation of BER performance of LoRa modulation when

affected by different types of ISM interference. Also,

authors quantiﬁed the loss of LoRa signal reception

sensitivity as function of SINR.

Furthermore, (Lauridsen et al., 2019) presents per-

formance evaluation of a narrow-band wireless Inter-

net of Things technology operating in the 865-868

MHz band. They evaluate the impact of RFID on

a narrow-band wireless Internet of Things technol-

ogy by applying the measured in-band power as in-

terference in a block error rate probability simulation.

Their measurement results showed that deploying a

wireless IoT technology in the 865-868 MHz band

will suffer from areas blocked by interference due to

the different transmit power and duty cycle restric-

tions in the European regulations.

In (Haxhibeqiri et al., 2018), authors showed the

evaluation of the impact of other sub-one GHz tech-

nologies (SigFox, Z-Wave, IO Home Control) on

LoRa technology. Their conclusion was that if the

interference starts during the preamble time, losses

can be as high as 28% in case of SigFox interference,

while losses are reduced, or even tends to zero, if the

interference starts by the end of the payload no mat-

ter the interfering technology. Also, spreading factor

of LoRa technology is considered an effective tech-

nique in case of interference since results presented

state that there is slightly high losses (5 to 10%) if

SF=7 is used instead of SF=12 for the same cases.

In addition, different surveys (Zhang et al., 2018)

talked about IoT technologies that share the same

spectrum in license and unlicensed bands provid-

ing spectrum sharing solutions, interference models,

interference schemes and disadvantages of sharing

spectrum. In (Guo et al., 2012) they mentioned the

mutual interference between different technologies

that operates at 2.4 GHz ISM band and suggest ways

to resolve interference. Moreover, in (Reynders et al.,

2016) authors showed the impact on network perfor-

mance when any of long range networks are deployed

in the same area and discussed the main technologies

LoRa and Sigfox. Their main work relied on the ra-

dio environment maps of interference and competing

base stations that could be very useful to provide con-

text information for these wireless networks. In (Ma-

madou Mamadou et al., 2020), authors present a sur-

vey on techniques in different technologies used in the

5G era that take into account coexistence with other

technologies.

As a conclusion, we realized that there is a limited

number of papers with speciﬁc focus on what may

cause interference in the license-free ISM bands and

how it may be harmful to short range devices operat-

ing in the same spectrum. Moreover, we did not ﬁnd

any studies that have been done on the performance of

NF-S32002/A1 protocol which is published in 2014

as a new intelligent wireless transport communication

system.

The added value of our paper, with regard to the

state of the art, is to measure the signal activity and

power levels of the transmission link established be-

tween remote-control transmitter and the pedestrian

trafﬁc light receivers. The results analysis are mainly

focused on the SINR threshold level in case of IoT

interfering nodes which are LoRa, Sigfox and En-

ocean which are typically spread all over in Clermont-

Ferrand near the trafﬁc light repeaters.

WINSYS 2021 - 18th International Conference on Wireless Networks and Mobile Systems

3 DESCRIPTION OF

NF-S32002/A1 PROTOCOL

Trafﬁc lights, in accordance with the decree of Jan-

uary 15, 2007 on application of decree no.2006-

1658 of December 21, 2006 relating to the techni-

cal requirements for accessibility of roads and pub-

lic spaces (French ministry of transportation and sea,

2007), include equipment allowing visually impaired

people to know when it is possible for them to go

through lanes. The associated R12 pedestrian signals

must be supplemented by tactile or sound devices.

These devices comply with NF-S32002/A1 standard

(Leroux, 2009). They are considered as essential el-

ements for optimal operation of pedestrian lights al-

lowing the blind and visually impaired to cross the

road knowing the state of the trafﬁc light.

There are approximately 5000 sound beacons pub-

lished in France by Okeenea according to the ministry

of ecological transition of France. The R12 signals,

commonly called “pedestrian” ﬁgures, are made up of

two bright silhouettes, one moving for green, and the

other motionless for red. They must include a device

allowing blind or visually impaired people to know

the periods when the crossing is allowed. This de-

vice can be tactile or based on sound. When equip-

ping their pedestrian lights, many cities provide re-

mote controls free of charge to residents who need

them. Messages are emitted by these speciﬁc de-

vices, which operate permanently, semi-permanently,

by manual activation or by remote control activation.

These tactile or sound devices are always associated

with an R12 signal. For example, when the devices

send a tactile message, it is characterized by the emis-

sion of a vibrating or rotating movement over a suit-

able surface for the duration of the signal green R12

associated. Today, the touch is abandoned in favor

of sound, which responds much better to the expec-

tations of users. Technical characteristics of repeater

devices sound of R12 signals are speciﬁed in standard

NF-S32002/A1.

The EO-Evasion NF-S32002/A1 remote control

is an essential complement to all sound accessibil-

ity equipment. It enables all sound devices intended

for the visually impaired to be triggered: pedestrian

lights, sound beacons, posts, passenger information

terminals, etc. These devices meet accessibility stan-

dards and, by transforming visual information into

sound, improve the mobility and safety. The EO-

Evasion universal remote control is therefore an es-

sential tool for a visually impaired person, just like

the white cane or guide dogs.

NF-S32002/A1 standard is implemented on the

trafﬁc light repeater devices using the radio frequency

868.3MHz. This standard puts an end to prejudicial

situation for users, since the pedestrian had to have

up to three remote controls on them to be able to acti-

vate all the lights. After upgrading the remote-control

standard one remote control can activate all sound

beacons both for safety of pedestrians crossing and

orientation around and inside the buildings.

If the sound repeater device is activated by a re-

mote control, it must be able to receive an order which

transmission characteristics are described in this next

paragraph. This activation is called Interoperability

Transmission Link. It should be noted, however, that

this mode of activation is not exclusive; other means

of transmission may be added to it. For example, the

devices may have several other radio activation. The

general characteristics must comply with the RTTE

1999/5/EC directive and the harmonized standards:

ETSI EN 300 220-1, ETSI EN 301489-3, and ETSI

EN 300 220-2 (ETSI, 2019).

The transmission frequency must be centered on

868.3MHz, the signal will be transmitted in ampli-

tude modulation at a power less than or equal to 25

mW. When the pedestrian presses the button of the

remote-control, an RF signal is transmitted with am-

plitude modulation (ASK) more speciﬁcally with On-

Off shift keying (OOK) which is also similar to binary

shift keying BPSK from the point of view of channel

occupancy. This Remote control is one of the SRDs

that obey to duty cycle limitations and a maximum

transmission power equal to 14 dBm. The code mes-

sage that is sent through the RF signal consists of two

parts: a header and a code of 24 bits. In fact, there

is no MAC layer or a medium access technique or

even a data link layer. Whenever the person press

the button, an RF signal is transmitted by the physical

layer. The bandwidth used in this technology varies

between 100 and 300 kHz according to the manufac-

turer of the remote-control device, and the transmis-

sion rate or the bit-rate varies between 2 and 32 kbps

(Report ITU-R SM.2153-2, ). NF-S32002/A1 stan-

dard is designed to operate in France. Nevertheless,

the outcome of the study presented in this paper can

be applied to other protocols that operate at the 865-

868 MHz frequency band.

4 SIMULATION AND

EVALUATION RESULTS

All the presented results are done using a simulator

tool from the manufacturer ATDI which is HTZ com-

munications. The main functions of this simulator

helped us calculate interference levels by using all the

required tools to make full analysis of outdoor and in-

Simulation-based Study of Interference Impact in ISM Bands in Smart Cities: Connected Trafﬁc Light for Visually Impaired People

Use-case

door coverage by signal penetration, loss calculation

and point to point network analysis. Based on a real

city map, HTZ is able to provide precise signal prop-

agation impact on the received signal strength taking

into account various types of automatically detected

obstacles in the map.

In this paper, our basic analysis is to interpret the

performance of NF-S32002/A1 protocol that is lo-

cated in Clermont-Ferrand speciﬁcally in trafﬁc light

receivers. We have done different interference sce-

narios to test the received signal quality by interpret-

ing the level of SINR which is the main output of the

simulation in addition to the Received Signal Strength

Indication (RSSI) which describes the received power

level. Each scenario differs from the other by the ran-

dom distribution of the interfering IoT nodes (LoRa,

Sigfox, and En-Ocean) that are located near the traf-

ﬁc light receiver. These simulations describe a real

deployment case located in Clermont-Ferrand city.

HTZ uses the satellite cartographic maps that have

the same projection of map layers including all ob-

jects built in the city such us the buildings, streets,

the civil surroundings, and every object placed in the

area where the trafﬁc lights are located and the sce-

nario takes place. This simulator has a speciﬁc tool

that takes into consideration the real environment pa-

rameters that could affect the signal propagation and

penetration, such as obstacles, type of buildings, ma-

terial permittivity and conductivity, etc. All of these

essential tools of the simulator provide a deterministic

propagation model that is used into any phase of sim-

ulation. This deterministic model includes calculates

the generation of propagation losses due to (diffrac-

tion, absorption, ducting, reﬂections) attenuation. Be-

fore any simulation, the conﬁguration settings for the

base stations and their distributed subscribers must be

conﬁgured according to the global standards. In this

case, trafﬁc light receiver is conﬁgured as a base sta-

tion in order to receive RF signal from the remote-

control handled by pedestrians.

In the real deployment, the trafﬁc lights are

equipped with omni-directional antenna pattern of a

standard gain for SRDs approximately equals to 2.15

dBi. We mainly show the beneﬁts of using a direc-

tional antenna in this kind of use cases where the cov-

ered zone of the receiver is more or less well identi-

ﬁed and situated in a known direction. Thus, all the

results are evaluated according to these two antenna

patterns in order to compare the impact of interfer-

ence on the trafﬁc light receiver. Both antennas have

the same gain, only the directivity changes. In the

case of directional antenna, the directivity would pro-

tect the receiver from interference coming from nodes

positioned outside the covered zone. For a given user

at a speciﬁc position in the coverage of the main lobe,

the received power for both antennas would be the

same. Figure 1 shows the horizontal (on the left in

red) and vertical (on the right in blue) patterns of om-

nidirectional and directional antennas that we used in

our simulation scenarios.

Figure 1: Omnidirectional vs Directional antenna patterns.

The different interference cases that we simulated

are the following:

• Scenario 1: the position of the pedestrian is at a

typical distance of 25 meters from the trafﬁc light

where the interfering nodes and their base stations

are distributed randomly in this area. The number

of the interfering nodes varies from 25 nodes to

150 nodes.

• Scenario 2: we located the pedestrian at a variable

distance from 5 meters to 30 meters from the traf-

ﬁc light receiver. The interfering nodes are gen-

erated according to two different densities, 1 node

per square meter, and 1 node per 10 square meters.

• Scenario 3: the pedestrian is in front the trafﬁc

light receiver at a minimum distance of 5 meters.

A single IoT interfering node is placed at 5 me-

ters also from the trafﬁc light. The distance of the

interfering node is increased to reach 15 meters

from the trafﬁc light.

• Scenario 4: The last interference case describes

the case where a large number of indoor IoT sub-

scribers situated inside smart buildings that are lo-

cated at 10 to 20 meters from the trafﬁc light re-

ceiver.

WINSYS 2021 - 18th International Conference on Wireless Networks and Mobile Systems

4.1 Scenario 1

We added subscribers around the position of trafﬁc

lights with a number varying between 25 and 150

nodes. Then, we assigned an omni-directional an-

tenna as a pattern of the receiver and extracted the

needed results to interpret the propagation results.

The comparison is done by ﬁxing the pedestrian at

the same location but changing the relative IoT nodes.

The relative interfering nodes can be wireless sensors

placed in parked cars, buildings, people handling such

devices, alarms or any vehicle crossing by the street.

Second step was to change the pattern to directional

one with the same relative positions of all IoT nodes.

According to digital communications calculations

of the target SINR for NF S32002/A1 protocol, the

SINR objective to ensure a successful reception of

the modulated signal with a minimum sensitivity of

the receiver (-105 dBm) must be equal or higher than

14 dB. Figure 2 represents the ﬁrst simulation results.

We can realize that SINR value is higher for a direc-

tional antenna (SINR = 20dB and SINR = 4dB re-

spectively) when the number of interfering nodes is

equal to 25 IoT subscribers. As the number of nodes

increases from 25 to 50, the interference level of di-

rectional antenna achieves the target SINR but it de-

creases to reach 15dB while the omni-directional one

did not reach the objective SINR and its value reaches

-16 dB with 150 IoT subscribers where the directional

gives an acceptable value SINR = 10 dB. This means

directional antenna equipped at the sound beacons en-

hances the signal quality between TX/RX and pro-

vides a good tuning for the power received.

Figure 2: Scenario 1: SINR level with increasing number

of interfering nodes.

4.2 Scenario 2

The results in this scenario are divided into two parts.

In each case we have different distribution of IoT

nodes with respect to city area. High density sce-

nario refers to spreading 1 node per square meter.

Low density scenario refers to spreading 1 node per

10 square meters. In this simulation, the distance be-

tween the transmitter and the receiver (pedestrian and

trafﬁc light respectively) varies from 5 meters to 30

meters. Figure 3 shows the values of SINR for high

density scenario with 50 interfering nodes. SINR val-

ues for an omni-directional antenna starts with a value

of 17.94 dB till it reaches 5.44 dB at 30 meters. As

for the directional antenna, SINR level starts at 32.06

dB and drops to 16.91 dB at 30 meters. Hence, using

a directional antenna helps achieve target SINR lev-

els (above 14 dB), whereas this level is not achieved

with omni-directional antennas for a distance above

10 meters.

Figure 3: Scenario 2: SINR level with 50 interfering nodes

deployed at high density.

With low density scenario, 112 IoT interfering

nodes are spread one in 10 square meters. Simula-

tion results in are shown in Figure 4. SINR levels

vary from 30 dB to 18 dB when the distance increases

from 5 meters to 30 meters respectively. On the other

hand, with an omni-directional antenna, SINR levels

decrease from 14 dB to 4 dB as the distance between

TX/RX reached 30 meter.

4.3 Scenario 3

The results in this scenario show the high impact of

one individual IoT interfere located near the trafﬁc

light while a pedestrian is trying to activate the sound

beacon of the trafﬁc light standing 5 meters away. In

this simulation we moved the interfering node grad-

ually. Figure 5 shows the variation of SINR values

with respect to the increasing distance between the

Simulation-based Study of Interference Impact in ISM Bands in Smart Cities: Connected Trafﬁc Light for Visually Impaired People

Use-case

Figure 4: Scenario 2: SINR level with 112 interfering nodes

with density 1/10.

interfering node and the trafﬁc light. When the in-

terfering node is at 5 meters, the values of SINR are

-7 dB and -26 dB for directional and omni-directional

antenna respectively. This shows the unsuccessful ac-

tivation of NF S32002 protocol due to high interfer-

ence level from the IoT interfered but still has a lower

impact when a directional antenna is used on the re-

ceiver. In addition to that, notice that SINR increases

to achieve a target value SINR of 15 dB when the dis-

tance was 10 meters and remains constant as the dis-

tance increases to 15 meters. On the other hand, SINR

did not reach the target value in omni-directional case

and remains below the required value in all distance

sections.

Figure 5: Scenario 3: SINR level with 1 inteferer node.

4.4 Scenario 4

This interference case describes the absence of inter-

feres in the street close to the trafﬁc light receiver

but we generated a large number of indoor IoT sub-

scribers inside the buildings that are close by. The

pedestrian is at a ﬁxed minimum distance in-front of

the trafﬁc light (5 meters). HTZ Communication sim-

ulator supports powerful tools that allows us to gener-

ate indoor subscribers placed in speciﬁc ﬂoors in the

buildings, or any place in the clutter of the map like

roofs of buildings, roads, forests, etc. The objective

here is to determine the impact of a smart building

equipped with many IoT devices sharing the spectrum

of 868MHz band at the same time when a pedestrian

crossing the road. Figure 6 shows the google hybrid

map with the distribution of indoor subscribers. Re-

sults of SINR are also represented in the graph show-

ing the difference of values according to the pattern

of antenna used.

Results showed, in case of omni-directional pat-

tern used on the receiver, that the SINR value was

-1.55 dB which is way under the threshold values

while the directional antenna achieved an SINR of

24.5 dB. This simulation explains the inﬂuence on the

interference level due to these indoor distributed IoT

sensors in buildings that lead to increase the propa-

gation losses. This is of course a worst case scenario

study, but the rational behind it is that in the near fu-

ture more and more buildings will be equipped with

wireless IoT objects.

Overall, results of SINR levels show the need to

use directional antennas to mitigate interference from

IoT devices. Bare in mind that using directional an-

tennas will reduce the coverage area of the trafﬁc light

receiver. This is why a careful deployment and posi-

tioning of these receivers should be done taking into

consideration pre-deployment simulation results for

better estimation. We argue that in this paper we

showed how using a realistic radio systems and signal

propagation simulator helps better estimate possible

interference levels when deploying wireless commu-

nication systems.

5 CONCLUSION AND

PERSPECTIVES

Wireless technologies are being used in many appli-

cations in our everyday life. A big part of the wire-

less technologies used by these applications use un-

licensed ISM bands for data transmission. With the

limited resources in the ISM band, this creates cases

of interference that might render the application non-

operational due to data loss. In this paper, we focused

on the speciﬁc use case of connected pedestrian trafﬁc

lights. These trafﬁc lights are used to help the visually

impaired to be aware of the state of the trafﬁc light.

Pedestrians use a remote control that operates at the

868MHz ISM band in order to communicate with the

WINSYS 2021 - 18th International Conference on Wireless Networks and Mobile Systems

Figure 6: Screenshot of HTZ simulator showing the placement of IoT nodes inside the buildings near the trafﬁc light.

trafﬁc light.

We studied the impact of interference in the

868MHz band on these trafﬁc light receivers using

a state of the art RF simulator called HTZ. We pre-

sented many scenarios where possible interference

might occur due to the presence of near by IoT wire-

less devices using technologies such as LoRa, SigFox

or En-Ocean. We emphasized on the importance of

using a directional antenna on the trafﬁc receiver in

order to enhance the SINR level and resist against in-

terference. The downside of using a directional an-

tenna might lead to limited covered, thus, it is very

important to carefully adjust the position of antennas

and their direction.

In our future studies, we will troubleshoot some

connected trafﬁc lights deployed in Clermont-Ferrand

city. Some of these trafﬁc lights do not respond reli-

ably to remote controls. We will analyse the deploy-

ment of these speciﬁc trafﬁc lights and compare the

RF measurements to what we obtained in our simula-

tions in order to propose adequate solutions.

REFERENCES

Augustin, A., Yi, J., Clausen, T., and Townsley, W. M.

(2016). A study of LoRa: Long range & low power

networks for the internet of things. Sensors, 16(9).

Elshabrawy, T. and Robert, J. (2018). The impact of ISM in-

terference on LoRa BER performance. In 2018 IEEE

Global Conference on Internet of Things (GCIoT),

pages 1–5. IEEE.

ETSI, E. (2019). 300 220-1 V2. 4.1 (2012-01). Electro-

magnetic compatibility and Radio spectrum Matters

(ERM).

French ministry of transportation, equipment, t. and sea

(2007). French decree.

Guo, W., Healy, W. M., and Zhou, M. (2012). Impacts

of 2.4-GHz ISM band interference on IEEE 802.15. 4

wireless sensor network reliability in buildings. IEEE

Transactions on Instrumentation and Measurement,

61(9):2533–2544.

Haxhibeqiri, J., Shahid, A., Saelens, M., Bauwens,

J., Jooris, B., De Poorter, E., and Hoebeke, J.

(2018). Sub-gigahertz inter-technology interference.

How harmful is it for LoRa? In 2018 IEEE inter-

national smart cities conference (ISC2), pages 1–7.

IEEE.

Lauridsen, M., Frederiksen, F., and Rodriguez, I. (2019).

Feasibility of deploying wireless Internet of Things in

the unlicensed European 865–868 MHz band. In 2019

IEEE Wireless Communications and Networking Con-

ference (WCNC), pages 1–5. IEEE.

Lauridsen, M., Vejlgaard, B., Kov

acs, I. Z., Nguyen, H.,

and Mogensen, P. (2017). Interference measurements

in the European 868 MHz ISM band with focus on

LoRa and SigFox. In 2017 IEEE Wireless Communi-

cations and Networking Conference (WCNC), pages

1–6. IEEE.

Leroux, A. (2009). Directional guiding devices, or mark-

ers on pedestrian passages or sidewalks for blind and

partially sighted people.

Mamadou Mamadou, A., Toussaint, J., and Chalhoub, G.

(2020). Survey on Wireless Networks Coexistence:

Resource Sharing in the 5G Era. Mobile Networks

and Applications, 25:1749–1764.

Simulation-based Study of Interference Impact in ISM Bands in Smart Cities: Connected Trafﬁc Light for Visually Impaired People

Use-case

Minaudier, C., de Lyon, C. ’e. U., Le Bras, A., de Paris, V.,

Le Loc’h, J.-P., Leroy, P., Vagogne, X., Landier, S.,

and Borsu, M. (2006). R ’e p

e titrators of pi ’e tones

for blind and visually impaired people.

Report ITU-R SM.2153-2. ITU-R: Technical and operating

parameters and spectrum use for short-range radio-

communication devices**.

Reynders, B., Meert, W., and Pollin, S. (2016). Range and

coexistence analysis of long range unlicensed commu-

nication. In 2016 23rd International Conference on

Telecommunications (ICT), pages 1–6. IEEE.

Saelens, M., Hoebeke, J., Shahid, A., and De Poorter, E.

(2019). Impact of EU duty cycle and transmission

power limitations for sub-GHz LPWAN SRDs: An

overview and future challenges. EURASIP Journal on

Wireless Communications and Networking, (1).

Zhang, L., Liang, Y.-C., and Xiao, M. (2018). Spectrum

sharing for Internet of Things: A survey. IEEE Wire-

less Communications, 26(3):132–139.

WINSYS 2021 - 18th International Conference on Wireless Networks and Mobile Systems