Leveraging Traffic Condition using IoT for Improving Smart City Street Lights

Full text

Leveraging Traffic Condition using IoT for
Improving Smart City Street Lights
Syeda Roushan Arshad
National University of Sciences & Technology
Islamabad, Pakistan
syedaroushan114@gmail.com
Aman Saeed
La Trobe University
Melbourne, Australia
amankamoka007@gmail.com
Vishwesh Akre
Dubai Women’s College
Higher Colleges of Technology
Dubai, UAE
vakre@hct.ac.ae
Hasan Ali Khattak
National University of Science and Technology
Islamabad, Pakistan
hasan.alikhattak@seecs.edu.pk
Sheeraz Ahmed
Department of Computer Science
Iqra National University
Peshawar, Pakistan
sheeraz.ahmad@inu.edu.pk
Zia Ullah Khan
Directorate of Science & Technology
Government of Khyber Pakhtunkhwa
Peshawar, Pakistan
ziaullah.05@gmail.com
Zahoor Ali Khan
Faculty of CIS
Higher Colleges of Technology
Fujairah, UAE
zkhan1@hct.ac.ae
Asif Nawaz
Department of Electrical Engineering
Higher College of Technology
Dubai, UAE
anawaz@hct.ac.ae
Abstract—This study introduces an adaptive street light
system based on sensor units that makes the system
cost efficient and reduces the power consumption across
highways on the basis of traffic conditions and intensity.
The main goal of the system is to develop a power saving
and efficient street light system and take a initiative towards
generating street light poles smart and intelligent using IoT.
Enormous use of street lights is consuming large amount
of electricity due to their uninterrupted operation all over
the night to manage traffic on daily basis. To minimize the
large consumption of electricity across highways there must
be a technological system that works smart and modify the
street light poles based on traffic conditions. This paper is
concerned with evolution and implementation of IoT based
smart street lights.
Index Terms—Detection, Sensor Unit,Dimming Circuit,
computation of energy saved, HTTP
I. INTRODUCTION
Modern Societies have slowly converged and have
adapted to use internet connected services and infras-
tructures. These infrastructures though seamless yet have
given huge benefits, we can use the internet connected
devices and services to connect and access everything in
the physical world. The applications of this integration
range from healthcare to transportation and building
infrastructures to industrial revolution. By connecting
cyber world with physical world almost every aspect
of human life can be automated using the internet and
giving them a unique identification. We can access and
control those smart things from anywhere in the world.
Due to urbanization and the modern world, the need
for electricity is increasing day-by-day, and most of the
electricity is wasted due to its inappropriate use [1]. It is
necessary to manage consumption of electricity and use
it in other productive businesses. In today’s world, about
50% of electricity is used for street light consumption.
In order to cut down this enormous consumption of elec-
tricity, powerful technology has to come up with smart
solution. An essential method for developing advanced,
efficient and energy saving systems for street light is
smart street light systems. This system intelligently
adapts the lightning levels based on traffic intensity. The
lightening levels will be adjusted by the system, using
sensors. This research work is based on principles of
using Internet of Things making the ordinary street lights
smart, adaptive and efficient [2].
The system consists of an ultrasonic sensor and real
time clock, which detects presence using echo principle
and schedules the system. Smart feature is added in the
system that it has been scheduled in a way that after
12AM the system turns the light poles dim as the traffic
intensity is less in mid night, whenever the presence
(vehicle) is detected the system turns bright and when
vehicle is out of the range of ultrasonic sensor system
then turn the pole dim and likewise, turn successive pole

bright and so on. The eternal feature is added that all the
pole values and active time of the pole is recorded and
stored on a web server firebase and displayed in the form
of visualization on user application from which saved
energy is calculated.
The paper also describes the efficiency of the system
as the poles are connected and communicating wire-
lessly using communication protocols of HTTP, while
calculating that how much of the electricity is saved and
consumed.The implemented system of street light is not
automated which leads to continuous operation of street
lights all over the night and because of that electricity
is continuously consumed which results in wastage of
electricity. By connecting poles wirelessly will help in
transferring point to point information to master node
and between poles, and defines the state of pole; active
or disabled, and take appropriate measures in case of
pole disabled or not working properly. This smart and
adaptive system is capable of saving energy along with
extended performance and reliability [3].
II. RE LATE D WOR K
Smart street lights are a project of smart cities. In
today’s world everyone and everything is connected to
the Internet in some way. Researchers have done much
research in this aspect. As smart street light is conserving
energy in the means of electricity, also it makes the
system more convenient and user friendly from its ability
to automate the switching on and off the lights. Various
algorithms have been proposed and used to improve the
efficiency of the system and many other are yet to be
proposed.
Yu-Sheng yang et al. [4] proposes a street lighting
management system consisting of cloud management
platform, and set of devices and sensors , and lighting
control function. The system is capable of providing
street pole data to the user in real time. The architecture
uses the container-based virtualization technology known
as Docker, to provide a strong and highly scalable solu-
tion to the deployment of the cloud and edge services.
Jain et al. [5] proposes an E-street system that works
on vehicle tracking, when vehicle is detected the system
will turn on the street light ahead of the vehicle and
which switches off the falling behind lights , if a vehicle
stops in between for longer time an alarm will beeps in
order to indicate the move the vehicle immediately to
save energy. Jain et al. used light senor and IR sensor to
detect vehicle and light. but we highlighted some issues
in the concept paper what if car broke down in he middle
of road.
Pattanaik et al. [6] proposed system in which decision
making for street light on or off is handled using ID3
algorithm. Their mechanism is that first street lights will
remain on for security reasons using the sensor. First
and last street light pole will remain on for the vehicles
using sensors for detection of vehicles. the reason is that
to reduce the number of sensors on street lights. Then,
prediction model comes in to identify the time slot where
peak traffic is available than usual. After that comparison
of previous data and current data will decide the time slot
at which street lights will remain ON in which most of
the traffic is available [7].
Mathew et al. [8] proposed an IOT based smart street
light controlling and monitoring using LoRa/LoRaWAN
network. This system keeps observing sensors working
condition and also control the on/off functionality from
a central point. This system works perfectly for both
outdoor and indoor lighting. Also it improves efficiency
of the system by sending notification in case of any
defect and emergency, also it extremely saves the electric
consumption by providing control over the sensors in
case of indoor appliances. The mobile app for the system
gives a user friendly and easily accessible platform to the
user.
C. Kruger et al [9] proposed monitoring smart street
lights faults monitoring over powerline using frequency
shifting key modulation, also they used LDR based light
sensor and LED based light intelligent control which is
built in power meter. Rest of this study involves PLC
which is a communication method implemented on an
existing power line. For smart street requirement narrow
band PLC are prefered which worked well with com-
munication for the system and overall system deemed
feasible for implementation.
Another paper proposes [10] a smart street lighting
system based on Brute-Force search algorithm which
reduces electricity consumption among street light sys-
tem. The system consists of pole controllers, segment
controller, power line communication transceivers, lux
meter with powerline interface, PIR sensor, WiMAX
modem, and monitoring and management software. This
system schedules the switching lamps and also dim each
pole using a controller software. Electricity consumption
is calculated using brute-force search algorithm. The
switching and dimming commands are passed to each
lighting pole through power lines. The pole controller
operates and apply the controller commands. Moreover,
if any fault is detected, it is reported to the controller
which is shared with the management system at the end.
Another paper [11] presented an optimal statistical
analysis of varying time based traffic for street light
system for energy optimization. The system uses LED
technology along with zigbee mesh network to provide
maximum energy efficiency. This system is implemented
in real time environment to validate the performance of
the system, which results that system is capable of saving
68 % - 82 % energy.
Related to above another paper pesented by lau et

al. [12] proposed a distributed Traffic-Aware Lighting
Scheme Management Network that enhances the useful-
ness of the streetlights, and by minimising their energy
consumption to improve efficiency of the system. This
system follows an approach that we are using in this
paper. Lau et. al demonstrated their system by comparing
their system to conventional system. BAsed on their
simulations and schedule their system proves to be 45
to 98% efficient. Their paper is highly relateable to this
paper in means of comparison with conventional system.
Chetna et al. [13], each pole has PIR sensor and
Zigbee devices that sense the information of vehicles
and pedestrians collected by the sensor array, and also
provides the services like telemetry, humidity and tem-
perature. Zigbee based street light control aims to reduce
human error in the running of street lights.
Intelligent street light system is presented [14] in
which they simulate the working of their low cost and
intelligent street light using IR and PIR sensors to predict
the energy consumption. Fujii et al. [15] propose an idea
of an autonomous distributed controlled light system, in
which street light poles turn on automatically before the
arrival of pedestrians or vehicles and dim itself whenever
there is no pedestrian or vehicle to save electricity.
III. PROP OS ED MO DE L
The system is implemented in such a way that its is
very easy to install on every pole by setting the sensor
network on each pole. Currently the system is monitored
by local governments which manage turning on and
turning of the lights in their perspective timings. Re-
placement of these street poles is very costly. Moreover,
The proposed systems does not require any management
to monitor this system as this is automatically controlled.
Low Cost The main goal of smart street lights is
to save electricity, and to save cost.The Network of
sensors and hardware is not very costly. Implementing
this adaptive street lighting system commercially will not
deeply affect economical and financial conditions.
Ease in updates Each unit of system is easily updated
according to the environment. The setup is designed
in such a way to behave intellectually according to
environments such as Residential areas, Markets and
busy areas.
Safe and secure The main goal of this adaptive
system is to keep people safe and save energy. This
system will be active after midnight when all the street
lights poles will turn dim. Keeping public safety in
mind, ultrasonic sensors are attached to detect any object
presence. If someone is hiding behind the pole to attack
pedestrians there will be sufficient lightning to view any
presence.
Interoperable There is no difference in seeing the
system as the driver and pedestrian will always see lights
turned on. Nobody will notice that lights turn off or that
the system is smart.
IV. PROPOSED SYSTEM
A. Real Time clock:
Real time clock (RTC) is a normal clock that runs on a
battery of 3.3V or 5V and keeps track of the time. In this
system RTC is used for setting schedule of street light
poles and also in scheduling with dimming and brighten
of the light poles according to the schedule.
B. Pole unit:
It consists of LED, RTC, motion sensor, the com-
munication device, such as HC 11 RF module, and the
controller. Each pole unit has these sensors attached with
it. It turns on for a few minutes under the circumstances
that a presence is detected in the range. Then, the pole
sends the message to other pole units. After sending
messages to other pole this pole will turns it self dim
to save electric power.
C. Sensor Unit:
The ultrasonic sensor is used to detect an object’s
presence using an echo mechanism. Detecting presence
will automatically turns street light bright.
D. HC 11 RF module:
This module acts as a communication and controller
module. It sends out messages to other poles using radio
frequency that presence is detected.
E. NODE MCU:
Node MCU has built-in Wi-Fi that act as master node
in this system which receives all the values from street
light poles and sends them to the cloud.
V. PROOF OF CONCEPT
In Figure 1, the basic architecture of the system is
shown. Sensors are connected to Arduino and Arduino’s
are connected to node MCU which shows the real time
working and data flow of the system. Our main goal
is to reduce the power consumption across highways
from street lights by dimming them when not in use.
DC converter is used to dim down the brightness of the
poles. As shown in Figure 2 the system works in a way
that there are RTC and motion detecting sensors, the
system is scheduled using RTC that from 6 PM to 12AM
midnight, the poles will remain bright. Another schedule
starts from 12AM till 6AM in which ultrasonic sensor
takes a part that all the street light poles turn dim after
12AM and when a car arrives in the range of sensor
that is, 4 meters, the light pole turns bright and when
vehicle is out of its range, the pole passes a message
to successor pole to turn itself bright and previous pole
turns dim using HTTP protocol.

Figure 1 explains the circuit of the master node.
One pole has an arduino Node Mcu which has built
in Wi-Fi , and that pole will act as master node in
this system.only one node will have node mcu. After
receiving the values from the other poles denoted as
slave nodes ,we calculated for how long each pole turned
bright, to evaluate total electricity (Energy) saved by
these adaptive smart street light systems. One master
pole in this system has Node MCU which receives
sensor readings and durations of being active from other
poles which are connected with simple Arduino’s and
sends the data to the web application which is based
on a firebase server. Once all the readings are saved on
server, web application fetch the data from server and
evaluates the electricity consumption by street light using
as follows:
P ower =Current ∗V oltage (1)
Power saved will be returned to the web application
showing how much electricity is used by the street lights
during their scheduled and peak hours routine for a day.
Once the hardware is integrated, the IOT platform
jumps in. Master node(Node Mcu) sends the deliberate
information to the internet connected web server (Fire-
base) for monitoring the values from each pole. Firebase
and web application is used for storing real time values
and allows the admin to access the website to monitor
data. The application allows the admin to turn the pole
active or disable.
VI. EX PE RI ME NT
Our system’s main module is the admin panel where
all the data and records are displayed with a proper
visualization of the total energy saved. Basically, the
software is dependent on the hardware part as the values
(time the pole remained on) are sent to the web server
from where all the values are fetched to the web app and
displayed in the proper format with look and feel. The
system is composed of RTC, HC 11 RF communication
module and ultrasonic sensor. In case of hardware, we
have built a prototype including 3 street light poles each
having one Arduino UNO. Each Arduino has an RF
module, RTC and ultrasonic sensor.
After the integration of hardware, node MCU is inte-
grated that act as master node and receives all the values
from street light poles that is the time of being in active
state and the amount of energy consumed. These values
are then transmitted to the firebase server console in each
row of each pole, which displays the total time according
to the schedule, as the poles are active from 6PM to
12AM with full bright lights and after 12AM till 6AM
they are dimmed and only turn bright when any presence
is detected by sensor unit. Firebase stores all the duration
of poles in an active state.
For the experiment, we kept on our poles of the
system from 6PM to 6AM as the current running system
of street lights does, and noted readings of electricity
consumption in a single day. On the other hand we
scheduled poles in our prototype to remain full bright
from 6PM to 12AM and turn dim after 12AM till
6AM as this study proposes, We saved readings and
compared both scenarios and calculated efficiency of our
system which came out 37%. Web application results are
also attached that shows the difference between these
scenarios in Figure 2.
A web application for admin is there to view all the
necessary activities and functions of lights and manage
them which are automated and just a few clicks away.
This same admin panel is responsible for automating the
street lights while other functionalities include:
A. Pole management:
In pole management, an admin can add a pole using
a web portal and after integrating the hardware the
street light pole starts sending its status to a web portal
describing active or disable which shows whether the
pole is working or not. After integration, if pole sends the
status “Active” then that shows the pole is now working
like other poles, and if the status is “Disable” then it
shows that pole is not integrated properly. It also allows
the admin to set or turn off the current schedule or set
the new schedule for the electric pole.
B. Compare Records:
In comparing records, the admin can view the results
in the graphical or pictorial representation which shows
energy consumption by system energy consumption be-
fore implementation of this smart system, to make out
the difference between local street light system and this
smart system.
C. Record Management:
One of the significant functions of our systems is
reports which includes all the data of total energy con-
sumed, total energy saved and total time of poles being
active. It also calculates energy and the power each pole
consumed on the basis of its time of being active. Work-
ing with bright lights and also with half brightness across
each pole gives the total energy (Result) consumed by
adding up all the power consumed by each pole.
VII. CONCLUSIONS AND FUTURE WOR K
We conclude that from the proposed system, we can
save up to 37% electricity as compared to the current
system of smart street light and it is also cost efficient
to be deployed anywhere. In future, some new features
can be added such as turning the poles bright based on

Fig. 1. Realistic picture of Master Node.
Fig. 2. Comparison of Results on User Application.
the speed of an incoming vehicle and detecting the type
of presence, e.g., if any animal sits below the street light
pole then the pole will turn on. If the dog sleeps under
the pole, the system will then work intelligently and
measure the distance from the object, if the distance is
constantly the same the system will turn the street light
from bright to dim.
REFERENCES
[1] S. U. Islam, H. A. Khattak, J.-M. Pierson, I. U. Din, A. Al-
mogren, M. Guizani, and M. Zuair, “Leveraging utilization as
performance metric for cdn enabled energy efficient internet of
things,” Measurement, vol. 147, p. 106814, 2019.
[2] A. Siddiqua, M. A. Shah, H. A. Khattak, I. U. Din, and
M. Guizani, “Icafe: Intelligent congestion avoidance and fast
emergency services,” Future Generation Computer Systems,
vol. 99, pp. 365–375, 2019.
[3] H. A. Khattak, R. Hussain, and Z. Ameer, “Internet of vehicles:
Integrated services over vehicular ad hoc networks,” in Interna-
tional Conference on Smart Cities, Infrastructure, Technologies
and Applications, pp. 61–73, Springer, 2017.
[4] Y. Yang, S. Lee, G. Chen, C. Yang, Y. Huang, and T. Hou, “An
implementation of high efficient smart street light management
system for smart city,” IEEE Access, vol. 8, pp. 38568–38585,
2020.
[5] S. R. Jain, D. Dayanand, C. Gowthami, and K. Punith, “Vehicle
movement street light controller system,” tech. rep., EasyChair,
2020.
[6] Y. Jagadeesh, S. Akilesh, S. Karthik, and Prasanth, “Intelligent
street lights,” Procedia Technology, vol. 21, pp. 547 – 551, 2015.
SMART GRID TECHNOLOGIES.
[7] M. M. Farooqi, H. A. Khattak, and M. Imran, “Data quality
techniques in the internet of things: Random forest regression,”
in 2018 14th International Conference on Emerging Technologies
(ICET), pp. 1–4, IEEE, 2018.
[8] J. Mathew, R. Rajan, and R. Varghese, “Iot based street light
monitoring & control with lora/lorawan network,” International
Research Journal of Engineering and Technology, 2019.
[9] P. Du Toit, C. Kruger, G. Hancke, and T. Ramotsoela, “Smart
street lights using power line communication,” in 2017 IEEE
AFRICON, pp. 1581–1586, IEEE, 2017.
[10] M. Mahoor, F. R. Salmasi, and T. A. Najafabadi, “A hierarchical
smart street lighting system with brute-force energy optimiza-
tion,” IEEE Sensors Journal, vol. 17, no. 9, pp. 2871–2879, 2017.
[11] G. Shahzad, H. Yang, A. W. Ahmad, and C. Lee, “Energy-
efficient intelligent street lighting system using traffic-adaptive
control,” IEEE Sensors Journal, vol. 16, no. 13, pp. 5397–5405,
2016.
[12] S. P. Lau, G. V. Merrett, A. S. Weddell, and N. M. White,
“A traffic-aware street lighting scheme for smart cities using
autonomous networked sensors,” Computers & Electrical Engi-
neering, vol. 45, pp. 192–207, 2015.
[13] C. Badgaiyan and P. Sehgal, “Smart street lighting system,”
International Journal of Science and Research, vol. 4, no. 7,
2015.
[14] Y. Jagadeesh, S. Akilesh, S. Karthik, et al., “Intelligent street
lights,” Procedia Technology, vol. 21, pp. 547–551, 2015.
[15] N. Yoshiura, Y. Fujii, and N. Ohta, “Smart street light system
looking like usual street lights based on sensor networks,” in
2013 13th International Symposium on Communications and
Information Technologies (ISCIT), pp. 633–637, IEEE, 2013.