[Type the company name]

Smart Street Light Using

LoRa with Multi-hop Linear
sensor network
[Type the document subtitle]
Indrajeet Singh, Praved Singh, Dharmesh Saini
Gurukula Kangari Vishwavidyalaya, Haridwar

2021
 Abstract waste electricity during unused hours just because of
manual controls of street lights. A smart street lighting
With the increase in population and the increasing system is a project of intelligent illumination control of
demand for road connectivity throughout the country, street lights to optimize the problem of power
the length of highways is increasing with direct consumption and the illumination of streets.
proportion. As per the reports of “MORTH”,the road
length of India is 59.97 lakh Kms as of 31st March 2019 Smart lighting has received a lot of global
and is the 2nd largest road network in the world. In this, attention recently as it is an evolution from human-
the national highway comprises 1.32 lakh km, with centric lighting with LEDs and or Organic LEDs (OLEDs)
nearly 40% of the total road traffic of India. The for dual-use, illumination, and communication process.
increasing road network directly increases the demand This technology aims to fuse drivers with sensing, control
for Street lights thereby, increasing the amount of algorithm, and wireless communications to deploy
power and investment costs. However different scalable lightning solutions able to work autonomously in
approaches are being proposed for making systems the IoT system. At its core, a smart lighting system is
energy efficient and upgraded with the latest being conceived as an adaptable lighting system with an
technology. The conventional systems are not worth objective to move from human-centric systems to
being implemented on such increased road networks, automated systems for energy and cost-efficiency.
as they are manually operated and maintained, which
thus leads to a huge amount of meaningless energy and
Related Work
manpower wastage. As the solution to these problems, The existing system consists of a centralized monitoring
we developed an automated IoT-based Smart Street center, an end node interfaced with sensors that may or
Light framework, especially for Highways. The proposed may not include a connection with the Internet. Most of
work shows the practical implementation of the smart the frameworks available today use Bluetooth, ZigBee,
street light solution, with the help of integrated and Wi-Fi technologies for wireless communication. The
sensors, micro-controllers, and a network to drawbacks they come with,are either they are costly or
communicate using LoRa technology, with low power short-range (ranging from 10m to 100m) or consumes an
and cost, resulting in a secure, cost-effective, and adequate amount of power. After getting a deeper look
efficient solution. The network part is realized with Low at street lightning networks it is observed that it doesn’t
Power Wide Area Network (LP-WAN) LoRa technology require high bandwidth to work properly. So a better
using Mesh topology resembling in less use of LoRa solution with low power consumption, long-range on the
gateways, lowering the cost of architecture. The whole cost of lower data rate can be developed for the street
system will be maintained and monitored through the lights.
cloud in real-time.
To counter these limitations, a communication
Introduction protocol of long-range with high security in the
transmission is needed. Thus, LoRa based street light
India is the world’s third-largest consumer and producer
management addresses the above requirement. In
of electricity and electricity is one of the basic needs of
addition, it is also cost-effective and reliable in
nowadays, but due to immense power consumption of
transmitting packets at lower power with a low data
the present street lighting system, it is essential to give
rate.
more attention to improve the efficiency and power
consumption by saving energy. Street lights being the So, the goal of this paper is to design an automated
most important and expensive responsibilities of smart lightning framework for Indian Highways to which
highway authorities, which account for about 40-45% of focuses on saving power, reduce the setup and
maintenance cost especially on highways of remote SF = log2
areas with lower traffic.
Where Rc is the chip rate, and Rs is the symbol rate.
Background
The value of SF ranges from 7 to 12, determining the
a) LoRa Technology
number of bits required to transmit the same amount of
LoRa technology, offering low data rate,with low data. Higher SF means that more bits are necessary to
power consumption and increased battery life to around send the same information. However, it is possible to
20 years, working with Mesh topology it offers scalability deploy the enddevice further from the gateway.
and fault tolerance and detection to the network. Each
Communication parameters are determined by data
device either can independently communicate with the
rate, which is given by
hub directly and work as a centralized monitoring system
or can communicate directly with various devices
DR = SF ∙ ∙ CR
present in the network to work as a decentralized
system. Its operational range is from 5km in urban areas Where DR is data rate and CR is code rate. In LoRa the
to 14km in rural areas, which is more than 50 times as of available choices of bandwidth are 125, 250, or 500 kHz,
previous technologies. Uses CSS (Chirp Spread Spectrum) and coding rate from 4/5 to 4/8. Usually, networks use a
in a frequency band of 868 MHz which is license-free i.e. 125 kHz bandwidth. These parameters are shared over
we can use this network without any license. It also the LoRa network between transmitter and receiver so
supports tracking applications without the use of GPS or that the modulation and demodulation can be
additional cause of power consumption which will help performed. The relation between SF, bandwidth, and
in fault detection and tracking. packet size is essential to determine the ToA (Time on
Air).

Time on Air (ToA) is the time required for a signal to

travel a signal from the transmitter to the receiver. It is
important to calculate it so that system-level design
constraints of ToA and the receiver sensitivity can be
effectively managed. Symbol duration is the time needed
to send 2SF chip at the chip rate, and it is given by,

Tsymbol= .

ToA is the sum of the preamble field’s transmission time

Fig: Range, Power consumption and Data rate of Different and the transmission time of the payload field. Preamble
Communication techniques time is given by:

CSS modulation uses Spreading Factor (SF) to spread Tpreamble= (npreamble + 4.23) * Tsymbol
information over frequency. The SFs are orthogonal to
each other and define the number of chirps per signal Where npreambleis the number of programmed preamble
symbol. Lower SF decreases the time on-air and energy symbols. In the payload duration, the number of symbols
consumption, increase the data rate but cannot be in payload nPL is calculated as:
achieved in longer ranges. SF is given by:
nPL= 8+max 𝑐𝑒𝑖𝑙 ((𝐶𝑅 + 4), 0 , topology for the LoRa network makes the system more
( )
reliable to be used. And to be cost-efficient
cost we will be
where PL is the number of payload bytes, H = 0 when the using a Partially-connected
connected mesh topology as shown in
header is enabled and H =1 when no header is present, Fig _. If any of the nodes get down then also the network
DE = 1 when the low data ratee optimization is enabled will run smoothly.
and DE = 0 when it is disabled, CR is the coding rate from
1 to 4(FEC codes 4/(CR+4)).

Knowing symbol duration and number of symbols in

payload, payload duration is given by:

Tpayload= nPL* Tsymobol.

Therefore, time on air is defined as: Fig: Partial Mesh Topology

ToA = [npreamble + 4.25 + nPL]*Tsymbol. The benefit of using mesh topology over start topology
is, this one gives more scaling options, longer range,
range and
Two packets with the same packet size, using the same
reliability against failure of the network.
bandwidth but different SF, have completely different
ToA. The SF, bandwidth, and packet size is chosen Proposed work
directly affect the throughput of the link. Th
The device
transmitting with low SF has a higher throughput than a Our proposed system is divided into three major parts,
p
device with a high SF value. end node, communication, and application interface as
shown in Fig. _.
Table 1 summarizes the relationship between the SF,
Throughput, and ToA considering a packet with different
payload sizes, 125 kHz bandwidth, and without the duty
cycle restrictions.

50 Bytes Payload Max Payload Size

SF ToA(s) Throughput Payload Max ToA
ToA(s) Throughput
7 0.113 3543.1 242 0.394 4913.7
8 0.205 1948.2 242 0.697 2777.6
9 0.369 1082.1 115 0.677 1358.9
10 0.698 527.8 51 0.698 584.5
11 1.478 270.5 51 1.479 275.9
12 2.629 152.1 51 2.793 146.1
Table 1: Relation between SF, throughput, and Time on Air(ToA)

b) Partial Mesh Topology

Fig: Architecture used in the proposed work
A mesh topology is a network setup where each
computer and network device is interconnected with one a. End node
another. This topology setup allows for most
It consists of a power-adjustable
adjustable LED array, the motion
transmissions to be distributed even if one of the
sensor, the communication module, and the
connections goes down. It is a topology commonly used
controller.The end
nd node has two different roles, first to
for wireless networks. This shift from star to mesh
adjust the brightness on account of the data received
through its sensors and generate a message, secondly to type for proper maintenance to be sent. Here the server
adjust the brightness based on the message received is having data of each pole with their real-time health
through the network from nearby nodes. It turns on for a that can be accessed any time through the application
minute under the conditions that a motion is detected in interface. For now, it consists of a LoRa module as a
the defined area by the sensors including its sensor. receiver to collect information from the street light
Then, it generates the message with its pole ID. It network and display it on screen.
reduces power under the condition that any motion is
not detected in the defined area.

Experimental setup

a. STM32F1038C8T6 Development Module

Fig: Circuit Diagram of the End node STM32 is a family of 32-bit bit microcontrollers
microcontroller which
integrated circuits by ST Microelectronics. It consists of a
b. Communication processor core, static RAM, flash memory, debugging
interface, and various other peripherals. Its key features
The message generated by the end node is br brought up to
include low power requirements, up to nine
the LoRa network with the help of the communication
communication interfaces, and up to t 80 fast I/O ports,
unit, the LoRa module present on the lamp. Now
which makes it quite versatile for industrial usage.
usage
depending on the type of message header (alert or
regular) the different lamps present in the defined area b. Motion detection Sensor
process the message. If the message is with a regular
header then lamps will collect the Pole Id of the message A diffuse
iffuse sensor is a type of photoelectric sensor.
source to calculate its respective
ective position to adjust its In diffuse sensors, the emitter and receiver form part of
brightness accordingly, and
d if the message is of alert type the same unit. The optical beams are either parallel
para or
then the module will be working as a router to help the slightly converging. The presence of an object in the
message reach the gateway so that it could be sent to optical field causes diffused reflection of the beam. The
the server. receiver detects the reflection from the object itself.
Suitable for a distance of about 50-100
50 meters
c. Application interface
c. SX1278 LoRa Module
As soon as a message will be received at the server end a
notification will bee generated for the maintenance This module is a transceiver that features the LoRa long-
department with the pole id, pole location
location, and the alert range modem that provides ultra-long
ultra range spread
spectrum communication and high interference The
he settings for the LoRa module,
module chosen to keep the
immunity by minimizing current consumption. It requirements
rements of SSL in consideration are shown in Table
operates at an operating voltage of 3.3V with an 2.
operating frequency of 433MHz in India and a
Parameter Value
programmable but rate of up to 300kbps. It can even
Center Frequency 433 MHz
send data signals up to a maximum distance of 10 km Bandwidth 125 kHz
which makes it quite efficient for long-distance,
distance, secure Coding Rate 4/5
data transfer. Spreading Factor (SF) 7
Output power 20 dBm
d. Light Dependent Resistor (LDR)

It is a special type of resistor having no polarity which Table 2:: LoRa module settings
can be used to sense light intensity. Its property to
Flow Chart
change its resistance based on the intensity of light is its
key working principle. Its symbol is similar to a resistor
with an extra inward arrow indicating the light signals.

Fig: Message flow on the LoRa network

Calculation / Observation

Now for transmission time, the Time on Air for LoRa

Fig: Components used in the proposed work module with the settings defined in table 2, i.e. CR = 4/5,
SF = 7, and BW = 125 kHz, can be calculated according to
[ ]. For a 51 byte payload it has the following value:

Ton Air≅ 100 ms

Concerning power consumption, the SX1272 datasheet

provides the following values:

Supply current in Receive Mode (Boost modality off)

IRX= 10.5 mA

Fig. a) End node b) Complete experimental setup Supply current in Transmit Mode (Boost modality off,
RFOP =+13dBm)
 ITX = 28 mA Results Discussion

Values do not consider the power consumption of Easy installation and extension:
extension Each unit can be
additional components like sensors, microcontroller
microcontrollers, installed one by one to the network by setting the
etc. parameters. The system is autonomous-distributed
autonomous
controlled.
ed. No host computer is needed.

Low cost: Only the partss of mass production are used.

Reduced number of gateway nodes.

Reliable:: With the use of mesh topology over start

topology makes the system more reliable.

Increased Range: Mesh topology has a higher range of

activity over star topology.

Easy update: The firmware of each unit can be updated

easily. The control algorisms should be developed for the
Fig: Observed Data from Arduino Interface
areas to be used in,, like a busy highway or a mountain
The graph shows how the modern smart street light is area with less traffic.
decreasing power consumption. To plot the graph, data
Self-diagnosis:: The worst event is that the light does not
was collected for trucks passing hourly
ly on a highway for
turn on when the pedestrian
destrian comes.
come Each unit records
24 hrs on a regular day. Considering the
the failures, in which the motion is detected in front of it
worstscenariowith
scenariowith each truck passes ata different time
without the advanced notification
tification from the other units
i.e. single truck per minute. The graph is between SSL
and circulates the alert message for that unit over the
switching time and OSL switching time during a 24hr
network if the problem sustains.
period, showing a constant switch-onon time of OSL and
varying switching times of modern SSL. The area Conclusion
difference between these two variables can easily depict
the difference between power consumption or the Street lights are the major contributor to energy
saving of power with modern SSL solutions. expenditure for any country, so there is a need to reduce
power consumption by introducing new schemes. The
main aim of the paper was to increase the efficiency of
Smart Street Light currently available by implementing
SSL with LoRa modules using a Multi-Hop Linear sensor
network as a partial mesh network. Concluding with
proper implementation and testing the setup of multi-
multi
hop linear sensor network of LoRa giving the desired
performance as expected.

Fig: Graph of Old Street Light Switching time against SSL switching
time for 24 hrs.
 References Italy. Sensors 2019, 19, 402.
https://doi.org/10.3390/s19020402.
1. F. Adelantado, X. Vilajosana, P. Tuset-Peiro, B. 12. Berto, R.; Napoletano, P.; Savi, M. A LoRa-
2. Martinez, J. Melia-Segui and T. Watteyne, Based Mesh Network for Peer-to-Peer Long-
"Understanding the Limits of LoRaWAN," in IEEE Range Communication. Sensors 2021, 21, 4314.
Communications Magazine, vol. 55, no. 9, pp. 34- https://doi.org/10.3390/s21134314.
40, Sept. 2017, doi: 13. Y.M. Jagadeesh, S. Akilesh, S. Karthik,
10.1109/MCOM.2017.1600613. Prasanth,Intelligent Street
3. F. Turčinović, J. Vuković, S. Božo, and G. Šišul, Lights,ProcediaTechnology,Volume
"Analysis of LoRa Parameters in Real-World 21,2015,Pages 547-551,ISSN 2212-0173.
Communication," 2020 International Symposium 14. Dizon, Eisley &Pranggono, Bernardi. (2021).
ELMAR, 2020, pp. 87-90, doi: Smart streetlights in Smart City: a case study of
10.1109/ELMAR49956.2020.9219028. Sheffield. Journal of Ambient Intelligence and
4. Higuera, J., Llenas, A., and Carreras, J., “Trends Humanized Computing. 1-16. 10.1007/s12652-
in smart lighting for the Internet of Things”, arXiv 021-02970-y.
e-prints, 2018. 15. Cotrim, J.R.; Kleinschmidt, J.H. LoRaWAN Mesh
5. Chaojie Gu, Rui Tan, and Xin Lou. 2019. One- Networks: A Review and Classification of
Hop Out-of-Band Control Planes for Multi-Hop Multihop Communication. Sensors 2020, 20,
Wireless Sensor Networks. ACM Trans. Sen. 4273. https://doi.org/10.3390/s20154273.
Netw 15, 4, Article 40 (October 2019), 29 pages. 16. Cong Tan, Duong & Kim, Myung-Kyun. (2018).
DOI:https://doi.org/10.1145/3342100 Reliable Multi-Hop Linear Network Based on
6. Fujii, Yusaku &Yoshiura, Noriaki&Takita, Akihiro LoRa. International Journal of Control and
&Ohta, Naoya. (2013). Smart street light system Automation. 11. 143-154.
with energy saving function based on the sensor 10.14257/ijca.2018.11.4.13.
network. 271-272. 10.1145/2487166.2487202. 17. NoriakiYoshiura, Yusaku Fujii and NaoyaOhta,
7. Yusaku Fujii, NoriakiYoshiura, Akihiro Takita, and “Smart streetlight system looking like usual
NaoyaOhta. 2013. Smart street light system with streetlights based on sensor networks”, 13th
energy saving function based on the sensor International Symposium on Communications
network. In Proceedings of the fourth and Information Technologies (ISCIT), (2013),
international conference on Future energy 663-637.
systems (e-Energy '13). Association for 18. Huang-Chen Lee and Huang-Bin Huang, “A low
Computing Machinery, New York, NY, USA, 271– cost and non-invasive system for the
272. measurement and detection of faulty streetlights”,
DOI:https://doi.org/10.1145/2487166.2487202. IEEE transactions on Instrumentation and
8. Jagadeesh, Y.M. &Akilesh, S. & Karthik, S. Measurements, Vol.64, (2015), 1019-1031.
&Prasanth,. (2015). Intelligent Street Lights. 19. B.Abinaya, S.Gurupriya and M.Pooja, "IoT based
Procedia Technology. 21. 547-551. smart and adaptive you lighting in street lights",
10.1016/j.protcy.2015.10.050. 2nd International Conference on Computing and
9. Sánchez Sutil, F.; Cano-Ortega, A. Smart Public Communication Technologies (ICCCT ’17),
Lighting Control and Measurement System Using (2017),195-198.
LoRa Network. Electronics 2020, 9, 124. 20. Philip TobiantoDaely, HaftuTasew Reda,
https://doi.org/10.3390/electronics9010124. GandevaBayuSatrya, Jin Woo Kim and Soo
10. JuhaPetajajarvi, Konstantin Mikhaylov, Marko Young Shin, “Design of smart LED streetlight
Pettissalo, JanneJanhunen and JariIinatti, system for smart city with web-based
“Performance of a low power wide area network management system”, IEEE Sensors Journal,
based on LoRa technology: doppler robustness, Vol. 17,(2017), 6100-6110.
scalability and coverage”, International Journal of 21. Jha Ashish K, Bababe Adam B and Ishan
distributed sensor networks, Vol. 13, No. 3, Ranjan, "Smart streetlight management system
(2017),116. using LoRa technology", International Journal of
11. Abrardo, A.; Pozzebon, A. A Multi-Hop LoRa Science and Research (IJSR), Vol. 6,
Linear Sensor Network for the Monitoring of (2017),2349-2352.
Underground Environments: The Case of the 22. Traffic Data Computation Method, Publication No.
Medieval Aqueducts in Siena, FHWA-PL-18-027 August 2018