Arabian Journal for Science and Engineering

https://doi.org/10.1007/s13369-020-05111-3

RESEARCH ARTICLE-COMPUTER ENGINEERING AND COMPUTER SCIENCE

An IoT-Based School Bus and Vehicle Tracking System Using RFID

Technology and Mobile Data Networks
Muhammad Wasim Raad1 · Mohamed Deriche2 · Tarek Sheltami1

Received: 15 May 2020 / Accepted: 2 November 2020

© King Fahd University of Petroleum & Minerals 2020

Abstract
The growing concerns of families about the safety and security of their children prompted a major interest in developing robust
systems that offer efficient tracking and monitoring of children commuting between homes and schools. With the emergence
of the Internet of Things (IoT) technology, in addition to Radio Frequency Identification (RFID), developing such systems
became feasible and cost-effective. In this paper, we present the design and implementation of a comprehensive low-cost
system based on IoT that allows schools, parents, and authorities to track the movement of children while in school buses or
being transported in private vehicles in real time. The system is based on off-the-shelf passive RFID readers that are installed
within buses, next to bus stations, and pick-up points at school entrances. We have implemented the system as an application
connected to a MySQL database deployed over the Heroku’s versatile cloud platform. The different stakeholders (parents
and school administrators) connect to the system through the Internet using a secure Java GUI. The parents can track their
children along the bus route in real time. The system is built to integrate children monitoring within the buses, the tracking of
the buses, and the tracking of private vehicles transporting children to/from school. The developed system provides additional
security features such as access to safety records of bus/vehicle drivers and infringements. A variety of reports for different
activities of buses and vehicles can be generated in real time at the school level. The system was tested successfully over the
King Fahd University of Petroleum and Minerals large campus which includes several schools (kindergarten, primary, and
secondary).

Keywords Smart transportation systems · RFID · Internet of things · IoT · Vehicle tracking · GSM · GPRS · GPS

1 Introduction sideration the need to provide a favorable environment for

children’s living and learning. Moreover, anecdotal evidence
Transportation is a vital sector that countries around the globe indicates that the bus stops around our King Fahd Univer-
try to manage efficiently especially with the growing num- sity of Petroleum and Minerals (KFUPM) university campus
ber of children and older kids commuting between homes and get very crowded during peak hours, resulting in immense
schools. Children’s safety has always been a major concern frustration for school children as well as their parents.
and a fertile area of research which gained a lot of attention The commuting of students from homes to schools and
across the globe. The Smart Cities paradigm takes into con- back has always been a major source of concern for par-
ents. Children often get on wrong buses and get off at wrong
B Mohamed Deriche stops. Bus drivers may not be able to identify all the chil-
[email protected] dren and thus will not know in time if a child is missing.
Muhammad Wasim Raad In addition to that, many families send a driver to pick up
[email protected] their children, which poses a challenge in guaranteeing that
Tarek Sheltami the child gets on the right vehicle. Ensuring safety and secu-
[email protected] rity of school children is of primary concern for the society
at large. Instances of kidnapping of school children on their
1 Department of Computer Engineering, King Fahd University way home or to school, and delay of school buses or vehi-
of Petroleum and Minerals, Dhahran, Saudi Arabia
cles due to road traffic, have put additional stress on families
2 Department of Electrical Engineering, King Fahd University and the society at large. Recent studies show that the num-
of Petroleum and Minerals, Dhahran, Saudi Arabia

123
 Arabian Journal for Science and Engineering

ber of missing children while commuting between homes determine/provide the real-time position of assets, resources,
and schools has drastically increased worldwide [1]. In this people, or anything of value with high accuracy and mobility.
context, researchers proposed IoT-based systems to track stu- The wireless technology used for communication between
dents in school buses using a combination of RFID, GPS, tags and readers can take many forms including RFID, WiFi,
GSM, and/or GPRS technologies. Moreover, different pre- Bluetooth, ZigBee, UWB, Infrared, and GSM. Each of these
diction modes were implemented for computing the arrival technologies is preferred for a specific application and has its
time of school buses [1]. Badawy et al. [2] proposed an IOT- pros and cons [15]. Among the above-mentioned technolo-
based system which gives access to parents and school admin gies, RFID systems are rapidly gaining grounds especially
to track children comfort and safety inside the school bus in in the area of intelligent transportation systems [16]. The
real time. authors in [16] showed that RFID-based systems exhibit
As alternative solutions, some researchers developed IoT- many advantages compared to other techniques, and that inte-
based systems using an android platform, GPS and Raspberry grating RFID in intelligent transportation systems constitutes
Pi controller to track real-time locations of public buses [3]. a promising solution. The versatile and advantageous RFID
However, GPS does not work well between high buildings technology has proved to be very useful in the field of intel-
and in tunnels. Other authors proposed combining RFID ligent transportation and particularly in enhancing driving
technology with fingerprint identification algorithms to send safety. In order to meet the requirements of low-cost and
notifications during the entry and exit of students to the school low-power consumption, intelligent transportation systems
bus [4]. Nevertheless, the fingerprint component adds extra are combined with RFID technology to provide real-time
cost to the system and is not suitable for students with skin monitoring of vehicles. It is worth noting that information
diseases. In [5], Dhanasekar et al. used RFID for tracking and fusing has been a very rich topic of research over a variety
monitoring students during their trip from (and to) homes of applications [17, 18, 19].
on college buses. The system included a speed feature that Other researchers, [20], proposed a comprehensive traf-
can be sent to parents (if needed) through GSM. However, fic management system based on RFID and data analytics.
using GSM to send messages to parents. Particularly in case These include detection of traffic violations; traffic manage-
of emergency or accidents, is not reliable as there are areas ment capabilities for smooth flow of traffic. Metlo et al. [21]
with no cellular coverage or the parents mobile might be highlighted in their research the importance of intelligent
off. Sutjarittham et al. [6] proposed IoT-based solutions for transportation systems based on monitoring the location of
tracking students’ attendance in classes as well as car parking vehicles for different applications including the reduction of
availability and bus-stop estimated waiting times. The work road traffic congestion. In other work, Ning et al. [22] noted
proposed by Sutjarittham et al., however, was solely intended that GPS is unable to acquire real-time information of vehicle
for a full-fledged highly computational solution for a smart location in some areas such as tunnels, built-ups, and high-
university campus setup. rise buildings, hence RFID when combined with GPS and
The evolution of the Information and Communication GPRS can provide a better alternative solution for vehicle
Technologies (ICT) through ubiquitous computing and IoT monitoring and tracking problems.
led to a plethora of intelligent systems. These systems offer As can be seen from the above, finding a comprehensive
robust, real time, and fast communication services for sharing solution which is both cost and performance efficient for
information and data over the cloud such as the scenario of school children tracking continues to be a very challenging
smart cities [7, 8]. As noted in [9], IoT technology is expected problem. Many of the existing systems limit their work to
to prevail in transportation systems in the very near future. either tracking within the bus or tracking the buses and private
By 2025, up to 75 billion devices would be connected over vehicles. Here, we present our own solution to the problem
the Internet [10]. by introducing a comprehensive workflow for tracking chil-
In [11], the authors conducted a comprehensive review dren within buses as well as tracking the buses and private
of IoT-based smart public transportation systems and tech- vehicles. The proposed system requires minimum Internet
niques focusing on safety in avoiding road accidents. In connectivity, works in real time, and is built using basic off-
[12], the authors, on the other hand, focused on compar- the-shelf components.
ing different IoT-based architectures for bus monitoring and Our main contributions in this paper are: (1) A compre-
management systems including RFID, GPS, GPRS, GIS, hensive framework for intelligent transportation of school
etc. Besides transportation systems, diverse enabling tech- children that includes tracking within buses and private vehi-
nologies have also been used to track objects, animals, cles, as well as tracking of buses and private vehicles along
people, etc. Such systems are broadly called RTLS (real- prescribed routes, (2) A cost-effective system for online
time location systems) [13, 14]. RTLS are a combination access using mainly the cloud network supplemented par-
of hardware and software used to monitor and continuously tially with the mobile data network and battery-less low-cost
and long range RFID industrial standard tags, (3) A sys-

123
Arabian Journal for Science and Engineering

tem with the ability to identify bus delays at the level of

bus stations giving peace of mind to parents and school
administrators alike, (4) An embedded safety feature related
to real-time monitoring of traffic records (and violations)
of drivers, (5) A scalability feature enabling the system to
include, on the fly, more students, more bus station points,
more features, and more services, all deployed over the cloud
and connected to an MySQL database, and finally 6. A suc-
cessful implementation and testing of the system as a pilot
on the university campus using off-the-shelf components.
The rest of the paper is organized as follows. In Sect. 2,
we present a critical review of state-of-the-art systems used
for tacking. In Sect. 3, we present the proposed system with
its different building blocks while in Sect. 4, we discuss the
details of the system in relation to hardware and software.
Finally, Sect. 5 discusses our testing protocol and experi-
mental results followed by a conclusion in Sect. 6.
Fig. 1 Basic operation of a typical passive RFID system [23]

2 Related Work

RFID is a simple wireless technology that utilizes radio fre- also considered as a preferred solution for monitoring within
quencies for data transmission. The data are stored on tags, tunnels and indoor bus stations where the GPS method fails
these tags can be either passive, active, or battery-assisted- to work and wireless sensor networks are costly and power
passive (BAP). The active and BAP tags both utilize batteries hungry.
that allow them to communicate over a longer range that In [25], authors proposed a children-tracking system based
can go up to 1 km for enterprise usage and over 2 km on RFID technology, where children carry RFID tags, and
for military applications. Unlike battery powered tags, pas- the system is responsible for locating the children playing
sive tags use the RF (radio frequency) signal received from in a theme park by aggregating the readings from a network
the reader to generate power and transmit/receive data. The of deployed readers. The proposed system has the advantage
RFID technology, particularly the passive RFID, is consid- of utilizing existing mobile devices in addition to reference
ered to be a low-cost technology for energy harvesting saving tags for localization of children. Nevertheless, the system
power compared to wireless sensor solutions (WSN) and the authors propose is highly computational intensive and
GPS-based systems. Passive RFID systems at UHF band costly. Feng et al. [26] proposed an RFID-based tracking
use battery-less tags, they have practically no life-time limit, system for buses which is combined with Zigbee wireless net-
and the coverage range is the highest for this type of tags work technology to achieve robust communication between
(7–15 m). As shown in Fig. 1, a schematic diagram of the vehicles, the platform system, and the dispatch monitoring
basic operation of a passive RFID tag that is powered by the center. Despite being able to realize self-organizing network
reader’s RF signal using a coil [23]. communication between nodes, Zigbee is considered power
The received power, Pr , at the level of the tag depends on hungry and consumes more power than passive RFID. In
a number of parameters including the transmitted power, Pt , [27], Habadi et al. used RFID technology to monitor the
the distance (in meters) between the transmitting and receiv- inside of buses and provide feedback if a passenger is left
ing antennas, d, the antennas’ gains, Gt and Gr , and finally, behind or trapped inside the bus as well as monitor the level
the transmission wavelength (in meters): of CO2 inside buses and control the opening of windows
accordingly. Other developed systems include attractive fea-
Pt G t G r λ2 tures for parents such as sending notifications to parents using
Pr  (1)
(4π )2 d n GSM [28, 29]. Finally, Arifa et al. [30] used a combination
of RFID, GPS, and GPRS technologies to monitor entry and
RFID provides a reliable form of automatic identifica- exit of students to school buses.
tion, which makes it a cost-effective technology to use for Based on the above review of the literature, we summarize
localization in indoor environments. Furthermore, RFID has in Table 1 the main technologies used for tracking objects and
favorable characteristics such as contactless communication, vehicles. The table shows that passive RFID is an attractive
security and non-line-of-sight (NLOS) readability [24]. It and cheap solution for tracking both objects and vehicles

123
 Arabian Journal for Science and Engineering

Table 1 Comparison of various vehicle tagging technologies tracking as well as tracking and monitoring children within
Identification Pros Cons buses. The system is developed in a way to minimize the use
system of the mobile data network as much as possible, and provide
complete transparency and access in real time. Moreover,
License plate Hard to forge Hard to implement
the system is implemented using simple off-the-shelf com-
recognition [32] (secure)
ponents. The developed system includes a number of safety
Line-of-sight is
needed features such as warning notifications in the case of children
One sensor per taking wrong buses or sleeping inside the bus. The system is
parking spot also developed to cover monitoring private vehicles in addi-
Passive RFID [33] Cheap Signal fading tion to buses with the added feature of web-enabled access
Easy to use Range and to the safety records of drivers.
interference issues Most of the literature either focus on tracking school
Secure children commuting using school busses with various tech-
Active RFID [34] Better detection Costly nologies, or merely tracking busses using GPS. To the best
than passive RFID of our knowledge, this is the first attempt to integrate cost-
Longer range Battery dependent effective and scalable solution using RFID for tracking
Higher security children as well as tracking busses and private vehicles under
IoT paradigm. The solution we propose works very well as
an upgrade to our smart university campus by mitigating the
especially when the considered distances are short; which in elongated waiting times at school and university bus stops
the case in our application. without incurring extra cost and with very minimal overhead
In Table 2, we list the different approaches and tech- using the existing information technology infrastructure. Uti-
nologies used for vehicle tracking systems. The table is an lizing Heroku cloud platform as a service adds outstanding
extension to the work in [21] and includes more parame- and novel security features to our IoT solution like Advanced
ters in the comparison. Each system has its own advantages Encryption Standard (AES) algorithm not only for security
and limitations. GPS-based systems, for example, provide but also great speed. The advantages it provides strong secu-
the best accuracy but are costly and consume more power. rity from attackers.
To reduce power consumption and cost, the table shows that
other systems like RFID and wireless network-based track-
ing systems provide better solution. However, we should note
3 The Proposed RFID-Based Tracking System
that wireless network-based tracking systems are complex
and require ongoing network connectivity while RFID-based
As stated in the previous section, our aim is to develop a
systems don’t require ongoing connectivity, are simple, and
comprehensive solution to the problem of school children
user-friendly.
tacking while commuting between homes and schools. Con-
Following our analysis of existing systems, we opted
trary to most existing solutions, here we discuss a framework
to use passive RFID technology, which is seen as a cost-
that considers tracking within buses as well as tracking of
effective option for our application of interest. More impor-
the buses themselves and even private vehicles transporting
tantly, the proposed RFID-based solution does not need any
school children. In what follows, we will start by presenting
network connectivity and is user-friendly compared to using
our overall system architecture. We then discuss the detailed
the Zigbee protocol which is very complex. Moreover, the
setup within and out of bus/vehicle environments. Finally,
proposed solution is shown to outperform GPS-based sys-
we summarize our design requirements and constraints.
tems in indoor environments (see Table 2).
Our main objective is to develop a low-cost complete
system to monitor children commuting between homes and 3.1 The System Block Diagram
schools, by tracking school buses at the level of bus stations,
private vehicles at the level of parking entrances, and individ- As seen in Fig. 2, the developed system revolves around three
ual children at the level of entry/exit of the buses. Literature main components: the bus stations equipped with readers and
showed that most existing systems either focus on tracking antennas, buses or private vehicles equipped with tags, and a
buses and vehicles, or tracking children within buses then data processing unit (DPU) at the school. Within the buses,
using the mobile network to transmit relevant information we have Nano readers that are used to acquire children data
about the children. Based on some earlier work on monitor- at pick up and drop off points. Figure 3 shows the main func-
ing children accessing and exiting the school bus [31], we are tional entities: the child carrying an RFID tag, the parent, and
proposing a complete system that combines bus (and vehicle) the administrator. These functional entities interact with each

123
Arabian Journal for Science and Engineering

Table 2 Comparison of various

systems used for vehicle Parameters GPS/GPRS-based RFID-based Wireless sensor Smart
tracking [19] tracking tracking network tracking phone-based
(Zigbee) tracking

Cost Costly Cost-effective Cost-effective Cost-effective

Power High No power No power High
consumption consumption consumption
Accuracy Accurate outdoors Less accurate than Less accurate than Accurate outdoors
in areas with no GPS (except GPS (Except in areas with no
high buildings indoors) indoors) high buildings
Installation On vehicle On vehicle Infrastructure On vehicle
location
Indoor/outdoor Good for outdoors Good for indoors Good for indoors Good for outdoors
environment only and outdoors and outdoors only
Need for network Yes No Yes Yes
connectivity
Scalability Not scalable Scalable Scalable Not scalable
Complexity High complexity User-friendly High complexity User-friendly
Safety/security Low High High High

Fig. 2 Block diagram of the proposed system

other while the school children are on the move to schools Fig. 3 Main functional entities in tracking school children
and back to homes.

3.2 Within and Out of Bus/Vehicle Setups

works cited. For every detected tag at the entrance (for morn-
For each of the children taking the bus, an RFID tag is ing travel from home to school) or exit (for afternoon travel
attached to his/her school bag which is read by antennas fixed from school to home), the acquired information is sent from
at the doors of the school bus. At the entrance door, the school the Nano reader to the Data Processing Unit (DPU) (located
bus subsystem triggers the RFID reader/antenna to read and at the school) using GPRS (see Fig. 4). The GPRS module
identify the child when the door is open. We use a Nano was used to provide the Nano reader a mean to connect to the
RFID reader from Sparkfun. The design choice is based on Internet. The reader is connected to the GPRS module using
the fact that this type of Nano reader is small and shown to be serial interface and uses UART protocol for data transfer.
well suited for prototyping applications and for deployment On the other hand, an Arduino script that runs on the GPRS
within a bus environment. In addition, the Nano RFID reader module, opens a TCP/IP socket and listens to data received
is easy to interface with Arduino microcontrollers and is not from the RFID reader. Once data is received, it sends the tag
a proprietary reader like the readers used in other research ID along with its payload to the database implemented in

123
 Arabian Journal for Science and Engineering

Fig. 4 Connectivity of the on-board school bus system

Fig. 5 Communication link between the vehicle and the reader at the
parking entrance

the cloud. The tags store representative information for each

child.
The Nano RFID sends wireless signals through the radio
frequency antenna at a certain frequency, and tags within
the electromagnetic field of the antenna produce an internal
inductive current. With this energy, tags communicate back
with the stored data to the reader. The Nano reader has a built-
in anti-collision capability to read 150 tags/s based on the
EPCglobal Gen2 protocol. Note that for children commut-
ing using private vehicles, the tag attached to the vehicle is
enough to track the child. Figure 4 shows the overall on-board Fig. 6 Vehicle tracking at the entrance of school parking
bus system including the Nano reader, GPRS, and commu-
nication with the cloud.
The bus stations are also seen as scanning points. Each bus application using Java language and also we used a MySQL
is equipped with an RFID passive tag tailored to work in harsh database that includes all the information of the tags attached
environments. The tags are identified in the system by their to the vehicles, buses, and enrolled children. As soon as a
unique ID (UID). When the bus approaches a given station, particular vehicle arrives to pick up a given child, the RFID
its tag is checked to estimate (or adjust) the estimated time of reader (in the parking area) reads the unique vehicle ID and
arrival (ETA) to the next station. Similarly, private vehicles sends a message to the school data processing unit so that
transporting children are also equipped with an RFID tag the school administrator informs the child that the driver has
which is read and identified upon arrival to the school parking arrived. Figure 5 shows the practical deployment of the RFID
area. readers at the entrance of the school parking area. Figure 6
The different groups of users can access all tracking infor- shows the RFID reader and antenna installed at the park-
mation live from a web-based application. These subsystems ing area of the school for tracking and identifying vehicles
will now be discussed in more details. transporting children. The reader is connected to the DPU
For our system implementation, we opted to use UHF with a database storing the secure information pertaining to
(Ultra High Frequency) tags. In comparison with low fre- registered cars and drivers authorized to transport children.
quency (LF) and high frequency (HF) bands, UHF RFID
systems have several advantages including fast reading,
enhanced information storage ability, superior reading range,
3.3 System Design Requirements
and are cost-effective [35]. In [35], D’Errico et al. discussed
a school bus tracking system using UHF RFID technology
After surveying current systems for bus/vehicle tracking for
with GPS to achieve excellent tracking performance.
schools’ related applications, we formulated the following
Similarly, we also have a tag attached to the private vehi-
requirements for our design:
cles transporting children. The tag (UHF OMRON) [36] is
attached to the front-end of the vehicle on the bumper. A pow-
erful antenna connected to a Sirit UHF reader is installed at • The system is able to identify the buses and their assigned
the entrance of the parking area. We implemented the system lines.

123
Arabian Journal for Science and Engineering

• The users can interact with the system, provide feedback,

and submit queries.
• The administrators can access the system in real time.
• The system should be scalable in terms of number of
schools covered by the system, number of buses/vehicles,
and number of children in the system, as well as other
factors.

4 Detailed Design
Fig. 7 The Sirit RFID reader with 3 antennas
The developed system consists of two main parts: software
and hardware. The hardware consists mainly of the readers
(inside the buses, at the bus stations, and parking area), as The data processing unit (DPU), which is located at the
well as the tags attached to each of the buses, vehicles, and school, is a gateway computer able to process and trans-
enrolled children. The software consists of two main com- fer large amounts of data at high speeds. It hosts the main
ponents: front-end and back-end. The back-end component application as well as are all management operations, and
includes the developed application, the database, and the dif- database.
ferent communication protocols. The front-end component
involves mainly the different website interfaces for users’
4.2 The System Software
management and tracking/monitoring tasks.
The system software consists of two main components: front-
4.1 The System Hardware end and back-end as follows.
Back-end component The application managing the sys-
We select the Sirit RFID reader, which is commonly used in tem is developed in C#. The different communication proto-
diverse industrial applications [37]. The reader is configured cols are also embedded within the application including the
to read and identify tags in the range of the deployed anten- EPC global gen2 protocol for acquiring information from the
nas. In Fig. 7, we show the reader with three antennas. The tags and sending it to the database.
Sirit UHF reader is used at every bus station (and parking For the database component, we choose the MySQL as
entrance), and it is connected to a linearly polarized antenna Database Management System (DBMS), and we deployed
with an RF power of 30 dBm. In this work, we have used a it over the Heroku’s online cloud platform. The database is
UHF antenna with coverage range of 7–15 m, which is rea- stored and deployed on the cloud, but a local copy is stored
sonable for tracking. The linearly polarized antenna offers a on the DPU for backup purposes. Heroku is a cloud platform
gain of 3 dB over similar-sized circularly polarized anten- that offers Platform as a service (PaaS), which allows build-
nas. Note that the RFID system operates at 902–928 MHz ing applications and rapidly deploying these online [38–40].
band frequency and uses the ISO 18000 GEN2 protocol. The It also allows integration of various add-on services offering
antenna uses radio frequency waves to transmit a signal for several options for scalability and functionality enhance-
activating the tags. Once activated, the tag transmits data ments.
back to the reader. Note that the readers (Sirit RFID Read- Front-end component The front-end is basically a web-
ers) used at the entrance of the parking area as well as at based GUI developed using ASP.NET web forms technology.
bus stations are different from the simpler version readers This technology minimizes the development time and pro-
with shorter distances (or so-called Nano readers) which are vides a consistent web application with the benefits of
installed inside the buses as discussed earlier. flexibility and compatibility with C#. As an example, we
The tags distributed to children are different from those display in Fig. 8 the parent interface which we developed in
on top of bus/vehicle tags. Children are provided with simple our earlier work [31].
UHF RFID tags of the adhesive paper-thin type. Such tags There are two modes for login, admin mode, and user
work with most common RFID readers and can be attached mode. The admin can login and enroll new children, par-
to practically anything we can imagine. The tags can be read ents, and link parents to children. This interface is dedicated
and be written to, and have a special kill feature for protection for school authorities (or admins) only. Admins have also a
against hacking. Each tag comes with a TID (Tag ID) that complete access to the lists of enrolled children and their par-
can’t be changed, but the tag has plenty of memory for writing ents in the database, as well the location of buses, and finally
and reading additional information when needed. the safety records of drivers. On the other hand, the parents’

123
 Arabian Journal for Science and Engineering

Fig. 8 Parent interface mode

Fig. 10 A student with his bag tagged enters the bus

Table 3 Probability of detection for varying transmit power

Probability of detection (m)

Power (dBm) .2 m .3 m .4 m .5 m .6 m .7 m .8 m

21 .90 .82 .78 .76 .69 .62 .60

24 .90 .85 .83 .80 .75 .72 .67
27 .91 .83 .84 .81 .77 .74 .69

Fig. 9 A typical testing scenario of the RFID system on bus

interface mode, Fig. 8, is designed to help parents monitor

the location of their children while commuting. Fig. 11 A sample web-based school bus schedule

useful in monitoring the level of safety of the system and

5 System Testing and Discussion taking remedial actions when needed.
We also carried a number of tests to analyze the effects
A proof-of-concept is developed and tested using a dedi- of tags orientation with respect to the reader’s antenna. Fig-
cated school bus within campus, as shown in Figs. 9 and 10. ure 13 displays the extreme positions of the tags (in terms of
The Nano reader within the bus is connected to an exter- orientation) with respect to the Sirit reader’ antenna. We also
nal antenna so that the reading range a distance covers up present our results in terms of the probability of tag detec-
to 70 cm, which is the approximate distance between the tion as a function of the angle between a sample tag and the
antenna center of the first passenger seat. Such a range is reader’s antenna (see Table 4). The probability was averaged
enough to avoid duplicate readings. We carried several tests over 10 possible readings and the orientations were changed
to determine the optimal transmit power of the Nano reader. between 0° and 90° (perpendicular position) in increments
The results are summarized in Table 3. The table shows that of 10°. The results show that the probability of detection is
the best probability of detection in our setup (around 70 cm acceptable for deviation angles of up to 60°. The results in
between the reader and the first seat) is achieved when the Table 3 and 4 are extracted from various experiments done
transmit power is set to around 24 dBm. in RFID Laboratory.
Figure 11 shows a scenario for two buses. It includes esti- It is important to mention that the proof-of-concept is ini-
mated times of arrival (ETA) and the route (between stations) tially tested in a laboratory environment before deployment
of the buses in real time. This information is accessible by on the bus. The testing in the laboratory started by adjust-
both parents and administrators. Figure 12 displays a web- ing the antennas based on their polarization to maximize the
based GUI showing the information pertaining to the drivers reading probability of the tags inside and outside the labora-
authorized to pick up specific children as well as the drivers’ tory environment. The setup is used to simulate a practical
records in relation to traffic violations. Such information is deployment scenario.

123
Arabian Journal for Science and Engineering

Fig. 12 The web-based interface for the school administrators

Table 4 Effects of orientation angle on tag detection probability

Orientation angle Detected tags Probability of detection

0º 10/10 1
10º 9/10 0.9
20º 8/10 0.8
30º 9/10 0.9
40º 10/10 1
50º 10/10 1
Fig. 13 Tag position wrt to reader’ antenna orientations 60º 8/10 0.8
70º 7/10 0.7
80º 1/10 0.1
Furthermore, numerous tag reading trials for tracking 90º 0/10 0
buses and private vehicles are carried under different tag ori-
entations and elevations (as discussed above). The Sirit RFID
reader is configured to work efficiently over the range of 7–15 An actual experiment was performed in a student parking
m (practical range for arriving vehicles to the parking area lot for detecting a tagged vehicle on the move as it approaches
and buses passing next to stations). its destination, utilizing the Sirit RFID reader and the rugged
To determine the optimal RF power of the Sirit Reader UHF passive tags. The test was successful in reading the
for our application, we changed the power setup between tag up to a speed of 80 km/h. Further testing was carried
the minimum transmit power of 10 dBm to its maximum on grounds using a dedicated school bus. The Nano reader
power which is 30 dBm. The testes were carried in increments inside the bus successfully collected the children informa-
of 5 dBm. For all setups, there were some cases of missed tion and sent it to the DPU using GPRS. While testing the
detection, except when the transmit power is set to 25–30 system, we faced a number of challenges. One challenge was
dBM. Based on these experiments, we opted to configure the the interference and reflections of metal objects with the bus
Sirit reader to operate at the maximum transmit power of 30 metallic structure. This challenge was solved by placing the
dBm. tag on the top of the bus and orienting the antenna horizon-

123
 Arabian Journal for Science and Engineering

tally with a slight angle upward. Two other challenges limited scope of the work can further be improved by incorporat-
the successful detection of RFID tags particularly in the case ing diverse machine learning tools for better tracking, more
of tracking children boarding the school bus: (1) the nature of flexibility, and additional useful features.
Omni-directional antenna of the tags, which requires proper
orientation, and (2) the human body shadowing. The RFID
reader fails to read the tag when the direction of the tag is
exactly perpendicular to the antenna reader. This causes null
areas in which the reader cannot detect the tag. However, this References
happens only for a short time duration as the child moves
toward or far from the antenna. 1. Jisha, R.; Jyothindranath, A.; Kumary, L.S.: Iot based school bus
tracking and arrival time prediction. In: International Conference
After conducting a number of trials, we have solved the
on Advances in Computing, Communications and Informatics
body shadowing issue by attaching the RFID tag to the child’ (ICACCI), pp. 509–514 (20017)
bag. For private vehicles, the proposed solution based on 2. Badawy, E.; Elhakim, A.; Abdulhmeed, A.; Zualkernan, I.: An IoT
RFID tags attached to vehicles is very much appreciated by Based School Bus Tracking and Monitoring System. In: Interna-
tional Conference on Education and New Learning Technologies,
the drivers. We also carried an informal survey with a cohort
Barcelona, Spain, pp. 5537–5546 (2016)
including administrators and parents to get their feedback on 3. Kumari, M.; Kumar, A.; Khan, A.: IoT based intelligent real-time
the system. Overall, more than 75% of the people surveyed system for bus tracking and monitoring. In: International Confer-
expressed satisfaction or high satisfaction (using the Likert ence on Power Electronics and IoT Applications in Renewable
Energy and its Control (PARC), pp. 226–230 (2020)
scale) with the developed prototype. The developed proof-
4. Ahmed A.: An intelligent and secured tracking system for mon-
of-concept has recently been discussed with the University itoring school bus. In: International Conference on Computer
administrators (Transportation Department) who made some Communication and Informatics (ICCCI), pp. 1–5 (2019)
suggestions to further improve the prototype and add new 5. Dhanasekar, N.; Valavan, C.; Soundarya, S.: IoT based intelligent
bus monitoring system. Int. J. Eng. Res. Technol. (IJERT) 7(11),
features so that it can be formally adopted across all schools
1–4 (2019)
within campus. 6. Sutjarittham, T.; Gharakheili, H.H.; Kanhere, S.S.; Sivaraman,
V.: Realizing a smart university campus: vision, architecture, and
implementation. In: IEE International Conference on Advanced
Networks and Telecommunications Systems (ANTS), pp. 1–6
6 Conclusion (2018)
7. Fahmy, A.; Altaf, H.; Al Nabulsi, A.; Al-Ali, A.; Aburukba, R.:
In this paper, we discussed the design and implementation Role of RFID technology in smart city applications. In: Inter-
of a comprehensive low-cost RFID web-supported school national Conference on Communications, Signal Processing, and
their Applications (ICCSPA), pp. 1–6 (2019)
bus tracking system, with a number of valuable features. The
8. Sharad, S.; Sivakumar, P.B.; Narayanan, V.A.: The smart bus for
proposed solution was tested on buses and private vehicles a smart city—a real-time implementation. In: IEEE International
transporting children between homes and schools. The sys- Conference on Advanced Networks and Telecommunications Sys-
tem was developed so that instances of forgetting children tems (ANTS), pp. 1–6 (2016)
9. Luo, X.-G.; Zhang, H.-B.; Zhang, Z.-L.; Yu, Y.; Li, K.: A new
at the school or children riding on wrong vehicles are min- framework of intelligent public transportation system based on the
imized. Two sets of RFID readers were used: one set for Internet of Things. IEEE Access 7, 55290–55304 (2019)
reading the children’s tags on buses and the second is for 10. Ikpehai, A.: Low-power wide area network technologies for
reading bus tags at the level of bus stations or vehicle tags internet-of-things: a comparative review. IEEE Internet Things J.
6(2), 2225–2240 (2018)
at parking entrances. The system sends instant notifications 11. Jalaney, J.; Ganesh, R.: Review on IoT based architecture for smart
with relevant data (about commuting children and the bus public transport system. Int. J. Appl. Eng. Res. 14(2), 466–471
itself) to the school DPU in real time. Parents can log onto (2019)
the system and monitor the details of their children locations 12. Hannan, M.A.; Mustapha, A.; Al Mamun, A.; Hussain, A.; Basri,
H.: RFID and communication technologies for an intelligent bus
in real time. The system database is managed by a dedi- monitoring and management system. Turk. J. Electr. Eng. Comput.
cated administrator and is deployed over the cloud for mobile Sci 22(1), 106–120 (2014)
access. Deploying the Mysql database on the Heroku cloud 13. Banach, M.; Długosz, R.: Real-time locating systems for smart city
offered an additional level of security. and intelligent transportation applications. IEEE 30th International
Conference on Microelectronics (MIEL), pp. 231–234 (2017)
A set of test case scenarios were considered to check the 14. (2019) Real Time Networks, Ultimate 2019 Real Time Location
efficiency of the system and user satisfaction. All tests sce- System (RTLS) Tech Guide. https://www.realtimenetworks.com/
narios passed successfully the essential design specifications blog/ultimate-2019-real-time-location-system-rtls-tech-guide
for a cost-effective tracking and monitoring system covering (2019). Accessed on Oct 28, 2020.
15. Halawa, F.; Dauod, H.; Lee, I.G.; Li, Y.; Yoon, S.W.; Chung, S.H.:
children, buses, and private vehicles. A number of extensions Introduction of a real time location system to enhance the ware-
are now being considered for full deployment within KFUPM house safety and operational efficiency. Int. J. Prod. Econ. 224,
then shared with educational institutions in the region. The 107541 (2020)

123
Arabian Journal for Science and Engineering

16. Pedraza, C.; Vega, F.; Manana, G.: PCIV, an RFID-based platform 29. Ambedkar P.; Babu, P.S.: Smart school bus for children transporta-
for intelligent vehicle monitoring. IEEE Intell. Transp. Syst. Mag. tion safety enhancement with iot. international Journal of Innova-
10(2), 28–35 (2018) tive Research in Computer and Communication, 5(7), (2017)
17. Kumar, T.; Gupta, S.; Kushwaha, D.S.: A smart cost effective pub- 30. Arifa, K.A.R.; Asha, K.R.; Amrutha, J.; Anju, P.: Smart school bus.
lic transporation system: an ingenious location tracking of public Int. J. Sci. Eng. Res. 7(2), 1036–1039 (2016)
transit vehicles. In: 5th International Symposium on Computational 31. Zaki, M.; Alhussein, K.; Aalquraini, A.; Raad, M.: IoT school bus:
and Business Intelligence (ISCBI), pp. 134–138 (2017) children safety. In: Presented at the Conference: Smart Cities Sym-
18. Mohandes, M.; Deriche, M.; Aliyu, S.: Classifiers combina- posium, Bahrain (2018)
tion techniques: a comprehensiveeq review. IEEE Access 6, 32. Anagnostopoulos, C.N.E.; Anagnostopoulos, I.E.; Loumos, V.;
19626–19639 (2018) Kayafas, E.: A license plate-recognition algorithm for intelligent
19. Al-Ani, A.; Deriche, M.: An optimal feature selection technique transportation system applications. IEEE Trans. Intell. Transp.
using the concept of mutual information. In: Proceedings of the Syst. 7(3), 377–392 (2006)
6th International Symposium on Signal Processing and Its Appli- 33. Miniaoui, S.; Hashim, K.F.B.; Atalla, S.: Car tracking system:
cations (ISSPA), pp. 477–480 (2001) availing RFID business value in reducing banks’ loss from car
20. Meghana, B.; Kumari, S.; Pushphavathi, T.: Comprehensive traffic loans. In: 2nd International Conference on Communication Engi-
management system: Real-time traffic data analysis using RFID. neering and Technology (ICCET), pp. 117–120 (2019)
in: International Conference of Electronics, Communication and 34. Mizuno K.; Shimizu, M.: Transportation quality monitor using sen-
Aerospace Technology (ICECA), vol. 2, pp. 168–171 (2017) sor active RFID. In: International Symposium on Applications and
21. Metlo, S.; Shaikh, F.K.: Vehical tracking techniques in ITS: a sur- the Internet Workshops, pp. 19–19 (2007)
vey. Int. J. Adv. Stud. Comput. Sci. Eng. 7(4), 32–38 (2018) 35. D’Errico, L.; Franchi, F.; Graziosi, F.; Rinaldi, C.; Tarquini, F.:
22. Ning, Y.; Zhong-qin, W.; Malekian, R.; Ru-chuan, W.; Abdullah, Design and implementation of a children safety system based on
A.H.: Design of accurate vehicle location system using RFID. Elek- IoT technologies. In: 2nd International Multidisciplinary Confer-
tron. Elektrotech. 19(8), 105–110 (2013) ence on Computer and Energy Science (SpliTech), pp. 1–6 (2017)
23. Bin Ibne Reaz, M.: Radio Frequency Identification from Systems to 36. (2020) Omron RFID Systems. (2020, http://www.ia.omron.com/
Applications. https://doi.org/10.5772/53284, ISBN: 978-953-51- support/guide/. Accessed on: 28 Oct. 2020.:
1143-6, IntTechOpen (2013) 37. (2020) 3 M Traffic Safety Systems, IN610 Infinity 610 User’s Man-
24. Alsinglawi, B.; Elkhodr, M.; Nguyen, Q.V.; Gunawardana, U.; ual. https://fccid.io/M4ZIN610/User-Manual/User-Manual-15554
Maeder, A.; Simoff, S.: RFID localisation for Internet of Things 67. Accessed on: 28 Oct. 2020.
smart homes: a survey. arXiv preprint arXiv:1702.02311 (2017) 38. Lee, B.-H.; Dewi, E.K.; Wajdi, M.F.: Data security in cloud com-
25. Pang, Y.; Ding, H.; Liu, J.; Fang, F.; Chen, S.: A UHF RFID- puting using AES under HEROKU cloud. In: 27th Wireless and
based system for children tracking. IEEE Internet Things J. 5(6), Optical Communication Conference (WOCC), pp. 1–5 (2018)
5055–5064 (2018) 39. Peña, M.A.L.; Fernández, I.M.: SAT-IoT: an architectural model
26. Feng, X.; Zhang, J.; Chen, J.; Wang, G.; Zhang, L.; Li, R.: Design for a high-performance fog/edge/cloud IoT platform. In: IEEE 5th
of intelligent bus positioning based on internet of things for smart world forum on internet of things (WF-IoT), pp. 633–638 (2019)
campus. IEEE Access 6, 60005–60015 (2018) 40. Anitha, M.; Nandhinik, K.: Hybrid Cryptographic Security system
27. Habadi A.A.; AbuAbdullah, Y.S.: Intelligent safety school buses in Heroku cloud. IJETIE 5(10), 106–120 (2019)
system using RFID and carbon dioxide detection. In: 1st Inter-
national Conference on Computer Applications and Information
Security (ICCAIS), pp. 1–7 (2018)
28. Fadzir, T.M.A.M.; Mansor, H.; Gunawan, T.S.; Janin, Z.: Devel-
opment of school bus security system based on RFID and GSM
Technonologies for Klang Valley Area. In: IEEE 5th international
conference on smart instrumentation, measurement and application
(ICSIMA), pp. 1–5 (2018).

123