Smart Car Parking System Solution for the Internet

of Things in Smart Cities

Wael Alsafery Badraddin Alturki Stephan Reiff-Marganiec Kamal Jambi
Department of Informatics Department of Informatics Department of Informatics Computer Science Department
University of Leicester University of Leicester University of Leicester King Abdulaziz University
Leicester Leicester Leicester Jeddah
United Kingdom United Kingdom United Kingdom Saudi Arabia
Email:[email protected] Email: [email protected] Email: [email protected] Email: [email protected]

Abstract—The Internet of Things (IoT) is able to connect bil- their cars, resulting in a high number of cars competing
lions of devices and services at anytime in any place, with various for a few unoccupied parking spaces and security for their
applications. Recently, the IoT became an emerging technology. parked cars. Moreover, in indoor car parks, most of the time
One of the most significant current research discussion topics on
the IoT is about the smart car parking. A modern urban city has is wasted on searching for unoccupied parking place which
over a million of cars on its roads but it does not have enough also creates traffic jam.The situation becomes worse when
parking space. Moreover, most of the contemporary researchers there are multiple parking slots in every parking lane. Another
propose management of the data on cloud. However, this method issue is the air pollution caused by cars in halt, be it in
may be considered as an issue since the raw data is sent promptly the outdoor or indoor car parks. In addition, most of the
from distributed sensors to the parking area via cloud and then
received back after it is processed. This is considered as an population use their own cars for traveling [3], which leads
expensive technique in terms of the data transmission as well as to more congestion in cities which in turn delays the process
the energy cost and consumption. While the majority of proposed of finding an unoccupied parking space for the rest of the
solutions address the problem of finding unoccupied parking drivers. An important issue to be considered is how to reduce
space and ignore some other critical issues such as information such data transmission cost and how to save energy as well
about the nearest car parking and the roads traffic congestion,
this paper goes beyond and proposes the alternative method. The as provide real-time information quickly - these issues require
paper proposes a smart car parking system that will assist users urgent resolution. The idea of smart car parking systems is
to solve the issue of finding a parking space and to minimise the recently attracting more attention in many countries owing to
time spent in searching for the nearest available car park. In the need for a new way of finding available parking places.
addition, it provides users with roads traffic congestion status. The IoT provides the capacity to deal with such challenges,
Moreover, the proposed system collects the raw data locally and
extracts features by applying data filtering and fusion techniques as it can be designed to capture sensors data for monitoring
to reduce the transmitted data over the network. After that, points of interest in smart cities.
the transformed data is sent to the cloud for processing and Recently, researchers have explored the potential use of the
evaluating by using machine learning algorithms. IoT in public transportation services and urban computing
Keywords Internet of Things (IoT); smart city; smart car [4]. Several models have been proposed to provide drivers
parking; Disruptive Car Parks; embedded computing; Cloud with real-time information about available car parking bays
data mining; Fog Computing; Distributed Data Analytics; nearby [5]. Additionally, some of them proposed solution to
collect and send the data to the cloud processing center which
I. I NTRODUCTION
determines the solutions and provides these back to the car
The rapid industrial growth in the world is reflected in in- parks. Only a few studies have been conducted where the data
creased number of cars on the roads globally [1]. It is expected about available car parkings on-street parkings was gathered
that the number of cars in the world will increase significantly more effectively. It was done by checking the road status if
from 841 million cars in 2008 to over 1.6 billion cars in there was a traffic congestion or not and if the parking location
2035 [2]. Nowadays, the shortage of available car spaces is was near the user. Adding on that ,some research papers
evident in many public places such as stadiums, market areas, proposed that [6], [7], [8] there is a need to develop a smart car
hospitals, shopping malls and airports, hence, governments are parking that indicates directly the availability of unoccupied
looking to improve their existing transportation systems and parking spaces in the nearest car parks to users while at the
infrastructures. However, the slow progression of city planning same time it makes users avoid the traffic congestion on the
has broaden the issue even more [3]. Finding unoccupied user’s way to the chosen parking. Optional choice could be
parking spaces is a common problem in most modern cities, added to make the web service display only free-of-charge
especially during peak times of various festivals. This problem car parking places to the user.
appears mostly in the modern cities; individuals come by The main contributions of this paper are as follows. First,
the data is collected from different distributed sensors in

978-1-5386-4427-0/18/$31.00 ©2018 IEEE

indoor parkings and on-street parkings. Second, the collected sensors data locally, displaying the availability of car parking
data from the sensors will be analysed and processed locally spaces, ability to guide the user to find the location of the
with the help of IoT devices. It is proposed that a real parking spaces. It makes the driver avoid traffic congestion
time processing for the smart parking data is extracted from as much as possible. Furthermore, the system provides the
the sensors. The data will be evaluated by using machine driver with the closest parking from the user current location.
learning algorithms, which in turn processes according to An optional driver feature provided by the system is that
predefined conditions. In addition, the system includes mobile if the driver wants only to show the nearest free of charge
phone application that lets users easily check the nearest car parking spaces from the driver, then the system will show the
parking with avoiding possible traffic congestion via Google available free of charge parking. In terms of communication,
API, which provides a real time reading of the traffic status. wireless communication is used between sensors and micro-
The cloud web service will collect the data from fog micro- controller devices based on technologies such as Bluetooth,
controller distributed devices that are near users location to as being a low cost option. Also, the communication between
start analysing and processing data. Then the data will be users and the cloud will be via HTTP protocol. There are
transmitted to users to indicate the nearest available parking several steps into data management in IoT, from collection to
which offers the lowest traffic congestion. Therefore, the user analysis and results presentation. In this part, we are going
will receive an immediate response from the cloud showing to represent the data filtering and fusion approach methods of
the number of available parking places represented in the map the system. Furthermore, Figure 1 shows the architecture and
with the less crowded roads from users existing location. functionalities of the system and details are explained below.
This paper consists of four sections. The first section is the
A. Overview
introduction section, the second section is discusses challenges
to overcome in the currently used system. In the third section, There are several steps to process the data in IoT including
the proposed approach is represented which demonstrates the collection, filter, fuse, preprocessing, storing and delivering the
overview of the system and the architecture design for the data [9]. The data filtering function filters the sensor data to
system side as well as for the graphical user interface design. reduce the amount of transmitted data. In addition, data fusion
The last section contains the discussion, conclusion and future function aggregates and integrates sensor data to provide
works. with more comprehensive information and thus produce more
accurate and meaningful information to users. Researchers
II. C HALLENGES have shown an increased interest in data fusion and filtering
There are a number of main challenges in smart car parking in the field of IoT. Data processing can be done at source, in
projects related to processing the received data from the
parking sensors including data fusion and filtering [9]. Data
fusion techniques allow to aggregate and integrate sensor data
from several sources to produce knowledge. Data filtering is
important to minimise as much possible the transmitted data
in the network [9]. However, where and how to conduct data
fusion and filtering are still open challenges. In-network level
data processing is important because the unused, unnecessary
and private data is not going to be transmitted over the
network. In other words, the continuously generated raw
sensors data will not be sent to the cloud, which is costly
and expensive. Another challenge is to identify the machine
learning algorithms or capabilities that might help to process
the collected data. Moreover, it is crucial to make all parties
provide an excellent functionality with each other in the real
time and avoid occurring of errors as much as possible.
III. P ROPOSED A PPROACH
This section starts by describing the structure of smart car
parking system which consists of three main parts: parking
part which includes sensor nodes in indoor car parking or
on-Street parking, micro-controller devices and smart car
parking facilities. The second part will be the cloud which
represents the mediator between the car parking and the user
mobile phone application. The third part is the user side
represented by mobile phone android application. This system
has the following functionalities consisting of processing the Figure. 1: Functionalities of Smart Car Parking System.
the network and in the Figure. 1: Functionalities of Smart Car d) The mobile node : ”Android is an mobile operating
Parking System cloud. The research to date focused on cloud system developed by Google based on Linux kernel and
processing rather than in-network level processing. It is clear primarily designed for touch screen”[13]. Android operating
that the cloud computing is important for enabling the vision system is a stack of components that is roughly divided
of IoT from several perspectives such as data storage and high into 5 sections which are Linux kernel libraries, Android
data processing power [10], but this can be enhanced by IoT Runtime, application framework and component. The smart
middle-ware which supports data management including data phone application provides users with an interface to interact
processing and real-time services to IoT applications [11] in with the device. The application provides users with the real
other parts of the network. It is important to know where time reading of the available spaces for the nearest parkings.
the data processing will be done most effectively. Therefore, Users will be also able to receive a real time reading for the
this paper considers that processing sensors’ data locally with traffic situation for each nearest parking from the user. An
middle-ware before sending to the cloud for further processing additional feature is the Text to Speech module which ensures
and storage is highly relevant for the internet of things and thus that the users do not have to look at their phones while driving.
largely overlooked. Figure 2 shows the android software design.
The main objective of the research paper is a) an approach
IV. R ELATED WORKS
to filter sensors’ data to discover the existence of vehicles on
a parking spaces by using wireless sensor network through Various parking systems have been done in the past. It is
localised processing, b) a middle-ware architecture which apparent that the car parking systems implemented in devel-
can manage the generated data from sensors to support the oped countries are done with advanced technologies deployed
approach, c) present a novel parking system with support of in the car park. This section reviews various works conducted
the multi-Edge-Fog which will minimise the traffic congestion on smart car parking systems.
by displaying free vacant spaces, ensuring that it shows the Some available car parking systems researches are sensor
nearest parking spaces to users, and also to avoid the traffic based [14], [15], [16] which generally processed the problem
congestion as much as possible. of locating an unoccupied car parking place. Similarly the
same aspects were observed [17], [18], [19], [20] which
may lead to a higher traffic congestion while other parking
B. Architectures
recourses remain available to use. Moreover, the problem with
a) The Sensor node: This IR sensor node is the lowest these systems are that they are unable to find out a solution
level of the smart parking system and it is placed in every for previously mentioned issues such as the nearest car park
parking space. The IR sensors are used to detect the obstacles.
They are very cheap and easily available on the market. It
uses a voltage of about 0.5 volts-0.25 volts and it reduces
program complexity. If any vehicle is detected at that time,
the IR sensors senses the output which is red and when there
is no vehicle at that time its output is shown as green.Then, it
sends the data to the middle-wear micro-controller IoT device
by using wireless communication.
b) The fog node: It is micro-controller device which is
considered as fundamental part of the system. It is functional
for collecting sensors data using wireless communication and
it does filtering on the collected data from the sensors and
processes the data locally. This step is important to minimise
transmitted data over the network and to save the energy [12].
Furthermore, it sends the processed data which contains the
number of free car parking spaces to the cloud using web
protocols such as HTTP or COAP for storing and processing
the data.
c) The cloud node: The cloud role will be to receive the
processed data from the fog node and then two main tasks are
to be done by the cloud. First is to process the received data
with help of Google API and mainly apply machine learning
algorithms on the data and store them. Second is to send the
information to a user’s mobile phone application. It is clear
that the processing power and storage capability of the cloud
is high. This power can be used even more effectively by using
the presented approach. Figure. 2: Software Design.
available and the roads traffic congestions. Hence, the main android application shows the user the nearest parking spaces
focus was in determining the occupancy status of parking and the road traffic status of intended parking.
spaces.
Some of the researches [21], [22] focused on information V. D ISCUSSION
system and guidance for parkings. Those systems provide
The smart car parking system provides way to save the
users information via some message signs in actual time avail-
energy and minimise the transmited data over the network.
able parking spaces for a specific zone.The total occupancy is
There are a number of advantages of this system. The first
gathered by using the sensors on the both entrance and exit
advantage is to help users to save time by providing the right
of the parking location. Moreover, in [22] result of the survey
direction of the free car parking space. It also helps to show
shows that there is a huge desire from the participants for
users many available parking areas close to users’ location as
more dynamic, accurate and personalised parking information
well as the traffic situation. Another positive aspect is that
at both before the trip and en-route. However, in the proposed
the system is automated which does not require any user
system, the information represented to the user on the map via
intervention. Additional feature in the android application is
helping of Google API will be via micro-controller device that
the Text to Speech module which will ensure that users do
will save the energy and reduce the amount of transmitted data,
not have to look at their phones while driving for the purpose
while the final results will be represented on the user mobile
of safety. Additionally, using in-network level data processing
application.
will increase the privacy by isolating the system from global
The work [23] proposed system finding available parking
world because the data processing and storage is done locally
spaces depending on the visual sensor network. In addition,
rather than sent to cloud for processing. However, if we would
there were wireless cameras installed in the local area to get
try to extend our approach to add some more features, that
the image of the parking lots which sent them to a centralised
would be such as reservation service for the available car
system for processing. They demonstrated promising outcomes
parking spaces. Moreover, it could be further extended to
by deploying the system on a visual sensor network and
support other mobile phone operating systems such as Apple
the test presented how good was their method in terms of
IOS and BlackBerry. Furthermore, it could provide users with
performance and low requirements. However, this system gen-
the information about parking spaces in wider zone in case that
erates a significant amount of data which is to be transmitted
a user is travelling and want to know more about the available
over a network to a centralised system that will increase the
parking spaces and the traffic situation in the destination. After
transmission cost of data and consume more energy.
discussing the advantages and limitations of the system, this
Moreover, the study [24] proposed a cloud-based car park-
system will be useful to provide users with information in
ing method to reduce the consumption of time and gas while
the real time to save users time and make them avoid traffic
finding car parking spaces. Additionally, they used simulation
congestion as well as to activate the Text to Speech feature to
tools to show their system architecture and they have imple-
maintain the safety of the user while driving.
mented a real scenario. They used wireless sensor network and
RFID to control the car park that produced a large amount
VI. C ONCLUSION AND F UTURE W ORK
of data which were sent to the cloud for processing. The
shortcoming of their study was that they did not consider the This study has proposed a smart parking system that en-
expensive method of sending the huge amount of generated hances the performance of saving users time to locate an
data. appropriate parking space and reduces the general costs for
Furthermore, the work in [25] proposed a car parking system moving to chosen parking space. The most obvious finding to
in IoT in the context of airport car parking which allows the emerge from this study is that we proposed a smart car parking
registered users to login to the system to check the free spaces system that will make ensure reduction of transmitted data
in the parking. In addition, GPS is used to show the exact through the network and saving the energy in the perception
available vacancy in the parking area to users using cloud layer. While in the application layer side is to save the user
server. However, all the processed data is transmitted to cloud time, avoiding traffic congestion, find available parking spaces
for processing then the results sent to the users. This means and to reduce cars gas emissions from drivers while searching
a lot of data is transmitted over the network which involved for the empty parking spaces.
high cost of data transmission. Furthermore, we showed the proposed approach of the
Finally, most of the existing approaches did not attempt system and its relevant methods. In addition, on the basis
to localise data processing by moving some decision making of this presented paper findings, continuous improvement
closer to the source of the data. This rersearch proposed a the and work on the remaining problems is continuing and will
data management approach to process sensor data locally in be approached in future papers. Future work will involve
order to avoid the high volume of data to be transmitted over implementing the proposed approach in large scales in the
the network. In that attempt, we used WSN to monitor the real world and test the system to check the results. Finally,
car parking spaces. Furthermore, the data filtering and fusion the future step would be developing a mobile application for
techniques are used to reduce the energy consumption and cost users to find available parking slots and guide them through
of data transmission in the proposed approach. In addition, the this application.
 ACKNOWLEDGMENT [22] Y. Ji, W. Guo, P. Blythe, D. Tang, and W. Wang, “Understanding drivers
perspective on parking guidance information,” IET Intelligent Transport
Wael Alsafery’s and Badraddin Alturki’s research are Systems, vol. 8, no. 4, pp. 398–406, 2013.
funded by Saudi Arabian Cultural bureau in London and their [23] L. Baroffio, L. Bondi, M. Cesana, A. E. Redondi, and M. Tagliasacchi,
scholarship is granted by King Abdul Aziz University. “A visual sensor network for parking lot occupancy detection in smart
cities,” in Internet of Things (WF-IoT), 2015 IEEE 2nd World Forum
on. IEEE, 2015, pp. 745–750.
R EFERENCES [24] T. N. Pham, M.-F. Tsai, D. B. Nguyen, C.-R. Dow, and D.-J. Deng, “A
[1] L. Wenghong, X. Fanghua, and L. Fasheng, “Design of inner intelligent cloud-based smart-parking system based on internet-of-things technolo-
car parking system,” in International Conference on Information Man- gies,” IEEE Access, vol. 3, pp. 1581–1591, 2015.
agement, Innovation Management and Industrial Engineering, 2008. [25] M. Suresh, P. S. Kumar, and T. Sundararajan, “Iot based airport parking
[2] B. Turner, “Organization of the petroleum exporting countries (opec),” system,” in Innovations in Information, Embedded and Communication
The Statesmans Yearbook: The Politics, Cultures and Economies of the Systems (ICIIECS), 2015 International Conference on. IEEE, 2015,
World 2012, pp. 70–71, 2011. pp. 1–5.
[3] C. Wenzhi and L. Bai, “A smart roadside parking navigation system
based on sensor networks for its,” in IET International Conference on
Wireless, Mobile and Multimedia Networks. IET, 2006, pp. 1–4.
[4] M. Handte, S. Foell, S. Wagner, G. Kortuem, and P. J. Marrón, “An
internet-of-things enabled connected navigation system for urban bus
riders,” IEEE internet of things journal, vol. 3, no. 5, pp. 735–744,
2016.
[5] I. Samaras, N. Evangeliou, A. Arvanitopoulos, J. Gialelis, S. Koubias,
and A. Tzes, “Kathodigos-a novel smart parking system based on
wireless sensor network,” in Intelligent Transportation Systems, vol. 1,
2013, pp. 140–145.
[6] R. Kannadasan, A. Krishnamoorthy, and N. Prabakaran, “Rfid based
automatic parking syste,” 2016.
[7] S. Hanche, P. Munot, P. Bagal, K. Sonawane, and P. Pise, “Automated
vehicle parking system using rfid,” Volume-1, Issue-2, 2013.
[8] K. T. Dorjee, D. Rasaily, and B. Cintury, “Rfid-based automatic vehicle
parking system using microcontroller,” 2016.
[9] M. Wang, C. Perera, P. P. Jayaraman, M. Zhang, P. Strazdins, R. Shyam-
sundar, and R. Ranjan, “City data fusion: Sensor data fusion in the
internet of things,” in The Internet of Things: Breakthroughs in Research
and Practice. IGI Global, 2017, pp. 398–422.
[10] A. Zaslavsky, C. Perera, and D. Georgakopoulos, “Sensing as a service
and big data,” arXiv preprint arXiv:1301.0159, 2013.
[11] M. A. Razzaque, M. Milojevic-Jevric, A. Palade, and S. Clarke, “Middle-
ware for internet of things: a survey,” IEEE Internet of Things Journal,
vol. 3, no. 1, pp. 70–95, 2016.
[12] B. Alturki and S. Reiff-Marganiec, “Towards an off-the-cloud iot data
processing architecture via a smart car parking example,” in Proceedings
of the Second International Conference on Internet of Things and Cloud
Computing, ser. ICC ’17. New York, NY, USA: ACM, 2017, pp. 37:1–
37:5. [Online]. Available: http://doi.acm.org/10.1145/3018896.3018932
[13] D. Gavalas and D. Economou, “Development platforms for mobile
applications: Status and trends,” IEEE software, vol. 28, no. 1, pp. 77–
86, 2011.
[14] J. Yang, J. Portilla, and T. Riesgo, “Smart parking service based on
wireless sensor networks,” in IECON 2012-38th Annual Conference on
IEEE Industrial Electronics Society. IEEE, 2012, pp. 6029–6034.
[15] H. Wang and W. He, “A reservation-based smart parking system,” in
Computer Communications Workshops (INFOCOM WKSHPS), 2011
IEEE Conference on. IEEE, 2011, pp. 690–695.
[16] J.-H. Moon and T. K. Ha, “A car parking monitoring system using
wireless sensor networks,” International Journal of Electrical, Robotics,
Electronics and Communications Engineering, vol. 7, no. 10, pp. 830–
833, 2013.
[17] K. Mouskos, M. Boile, and N. A. Parker, “Technical solutions to
overcrowded park and ride facilities,” New Jersey Department of Trans-
portation, Tech. Rep., 2007.
[18] M. Idris, Y. Leng, E. Tamil, N. Noor, and Z. Razak, “ park system:
a review of smart parking system and its technology,” Information
Technology Journal, vol. 8, no. 2, pp. 101–113, 2009.
[19] G. Revathi and V. S. Dhulipala, “Smart parking systems and sensors:
A survey,” in Computing, Communication and Applications (ICCCA),
2012 International Conference on. IEEE, 2012, pp. 1–5.
[20] R. Ranjini and D. Manivannan, “A comparative review on car park-
ing technologies,” International Journal of Engineering & Technology,
vol. 5, no. 2, pp. 0975–4024, 2013.
[21] T. Rajabioun and P. A. Ioannou, “On-street and off-street parking
availability prediction using multivariate spatiotemporal models,” IEEE
Transactions on Intelligent Transportation Systems, vol. 16, no. 5, pp.
2913–2924, 2015.