Finalreviewpaper.doc

A Review On Smart Parking Systems In Metropolitan Cities Using IoT Ankit Patel Department of computer science , B Tech Integrated NMIMS UNIVERSITY, MPSTME Mumbai, India [email protected]

Meghan Patil Department of computer science, B Tech Integrated NMIMS UNIVERSITY, MPSTME Mumbai, India [email protected]

Abstract— In today’s world Internet has become an integral and important part our daily life. Most of the things we do are connected to the internet is some way or the other. IoT platform enables us to implement wide range of applications. It makes use of ultrasonic sensors, controllers, and interpreters to make to make the overall system smart. One such application is parking system. Smart parking system can be implemented by the use of IoT technology which provides various services such as information about parking spaces, the real time availability and unavailability of slots. Increasing traffic and number of vehicles, results in mismanagement of parking spaces. One has to spend a lot of time to find a vacant slot. IoT technology allows us to overcome such issues by using various components which helps to make the parking system efficient. In this paper, we have reviewed various systems that have been proposed and analyzed each of them and solved some of the issues that were encountered in these papers.

Mihir Prajapati Department of computer science, B Tech Integrated NMIMS UNIVERSITY, MPSTME Mumbai, India [email protected]

One of the major problems faced by humans in this decade is finding an empty parking space. Researchers have found out that in INDIA, there are about 50,000 vehicles registered every day. According to a recent survey conducted by Times of India, it was found that on an average 685 cars are being registered on daily basis. Mumbai had nearly 32 lakh vehicles last November (Fig 1). So we can imagine the rate at which traffic and the number of vehicles is growing.

Keywords— IoT, Smart Parking System, Smart City

I. INTRODUCTION Internet of Things is a concept where physical objects are connected to the internet. The term internet refers to scenarios where network connectivity and real-time computing comes into the picture. It deals with physical objects like vehicles, smart thermostat systems heart monitoring implants, Wi-Fi enabled washers/dryers and other objects embedded with software and sensors. The concept of combining computers and sensors has existed for decades but the recent market trends has attracted the Internet of things closer to widespread reality. As the arena of Internet of Things is growing, the applications based on it are also growing. In this paper, we proposed an IOT based smart online parking system which provides an easy and comfortable way of finding a parking lot and saves time and money.

Fig 1: Number of registered vehicles in India Most of the provided parking systems are considered to be just Informative Parking System (IPS) rather than “a smart parking system”[1]. One has to spend a lot of time to find a parking area. And even after successfully locating one, he/she faces a problem to reach that area. Even after reaching the parking area the person takes a lot of time to locate the exact spot. In most of existing systems, the driver has to spend a lot of time driving around the parking area to find an empty slot which results in wastage of many valuable resources such as time, fuel and so on. The result of a survey conducted recently showed that in India, it takes about 8.2 minutes for driver to successfully park their vehicle (low traffic). However, this value may vary depending upon the density of traffic. In cities like Mumbai and Delhi, it may take around 20 min. The

majority of time that is consumed in the parking process is caused by: (A) at the entry point where the driver has to get the parking card. (B) time taken to locate an empty space. In the previously proposed papers for Smart Parking System, the problem that was faced was that, once a user books a slot through the mobile application, the details of the slot and level number is provided to the user through the application. But, if accidently the user parks in some wrong location, then the entire system will be in error. In this paper, we propose a solution to the problems mentioned above by introducing an IOT based smart parking system which makes use of a mobile application to provide various smart services such as  Locating the available parking area.  Real time slot vacancy detection.  Navigation to a specific parking area.  Booking a slot (online/offline).  Detecting the car number plate  Recording the location of the parked car.  Calculating the fare and payment. The proposed system also provides an efficient way to ensure that the driver parks the vehicle in the allotted slot only. A. ISSUES 1. In our proposed system i.e, A Smart Parking System Using IOT, the problem that was faced was that, once a user books a slot through the mobile application, the details of the slot and floor number are provided to the user through the application. But, if accidently the user parks in some wrong location, then the entire system will be in error. 2. One of the most common problems faced by the drivers is that they are unable to find a vacant slot when they arrive at parking area. This problem may be caused due to various reasons like, peak hour parking etc. i.e, how the amount of time can be reduced for finding a vacant slot and efficiently park the vehicle? 3. What if multiple users are accessing the same slot at the same time during online reservation? 4. What if the user wants to park their vehicle only for a certain period of time? 5. How can we reduce the amount of energy consumed by the cameras installed at each parking slot? 6. What if the ultrasonic sensor fails to detect a vehicle present at the slot or detects the vehicle outside the slot area?

II. LITERATURE REVIEW A. Related Works In the previous works, several solutions were proposed in [2], [3], [4], [5], [6] for improving the existing parking system. Each of the proposed method is summarized is discussed below: [2] In this paper the Author has proposed a smart parking system which makes use of RFID (Radio Frequency Identification) and GSM (Global System for Mobile) technology which intends on providing a smart solution to deal with vehicle thefts. The main purpose of this technology is to make use of this wireless technology along with an application to solve the security issues that are encountered in an organization. The Author focuses mainly on the security aspects rather than focusing on the way the vehicles must be parked. Working of RFID AND GSM The main components of the RFID system are: 1. RFID interrogators 2. RFID labels 3. GSM kits The GSM kit is used to send SMS to an authorized user. RFID Readers and Tags are used to control the check in and check outs. RFID system consists of 3 components in combination of two  Transceiver (Transmitter and receiver) and antenna which forms RFID READER.  Transponder (Transmitter and responder) and antenna which forms RFID TAG. When the RFID READER emits a radio signal, it activates the transponder and the RFID TAG is read by the READER. There are two types of transponders, which correlate to the two major types of RFID tags Active transponder contains a battery that periodically transmits signals. Passive transponders do not have their own energy source and they rely on the energy given off by the reader [2]. Advantages  High security.  Eliminates the chances of human error.  Reduces operational cost. Disadvantages  If the RFID reader fails, the entire system will fail.  Check-in and check-out is only allowed in case of registered users.

[3] Author has proposed a smart parking system model which made use of an ultrasonic sensor which was fit in the ceiling of each car. Ultrasonic sensors are operated on the basis of echolocation. It emits a sound wave which hits the vehicle and reflects back to the sensor. The sensor is connected to a server which collects all the data and updates it in the application’s

LEDs which are used to indicate whether the slot is empty or not. Red color indicates that the slot is occupied and green color indicates that the slot is free to be utilized. The heart of this system is the embedded controller which interacts with the Raspberry pi and the Camera Module.



Step wise representation; 1.

Driver uses the mobile application to locate the nearest available parking area. The driver selects an appropriate option from all the listed parking areas 



Mobile application will send the requested information to cloud. It responds to the request and provides a map like structure of the nearest available parking areas. The available/unavailable slots are displayed in the application using graphics, colours and symbols along with the cost. 2.







 

The driver is provided with the navigation to the selected parking area. The driver follows the route specified by the application to get to the parking area. 3. The driver books the nearest available slot and makes the payment. After the slot is booked the user selects the amount of time they wish to park the car. According to the selected time, bill is generated and online payment is made by the user. 4. Driver is guided to the slot and the location of the parked car is recorded. The driver is navigated to their respective slots and after the car is parked, the notification to fill the license plate number is indicated on the application. Once the information is filled, the current location of the vehicle is noted and saved for later use. The driver is provided with an e-bill which can be accessed from the application. 5. The driver is notified 15 minutes prior to the slot time expiration. The driver then returns to the car and exits the parking area.

A notification is sent on the application along with location of the parked car 15 minutes prior to the timer expiration. If the user does not checkout within the allotted time, extra charges will be applicable. And finally, the driver tracks the vehicle position and returns to the location and leaves the area. To implement this proposed model, various components have been used. The system architecture makes use of following technologies and techniques:  Ultrasonic Sensors- it is used at each slot to detect the presence or absence of a vehicle. The data







obtained from the sensor is forwarded by the Raspberry Pi and updated in the mobile application. Camera- instead of asking the user to manually enter the license plate number into the application. The cameras can be installed which can extract the text from the number plate and store it in the database. This also provides evidence that a particular vehicle was present here at a particular time. Thus improving the security of the parking area. Raspberry Pi3- it controls the hardware data, which enables the sensor and camera to work together. It provides a connection to the cloud database. Mobile Application- it is software that is installed on the users Smartphone which enables them to access various services from the application. Cloud- used to store the data retrieved from the users and the sensors. It passes the information to and from the user and the sensor respectively. It is also useful to store the details of all the booking and payments.

Fig 2: Flowchart of various elements architecture

Fig 3: Ultrasonic sensor detection and LED color updation.



Fear of breakdown- there is always a fear that some hardware system may breakdown due to some reasons. Therefore, frequent maintenance is must.

[4] Author has proposed a system that serves as an e-valet

Fig 4: Camera to record the number plate text. When the driver brings the car close enough, the sensor is turned on and the image of the license plate is taken and the data is forwarded to the cloud.

Advantages/Disadvantage Advantages   













High parking efficiency- the users can find the best spots available with the least efforts. Saves time- the user no longer has to spend a long amount of time searching for an empty slot. Enhanced security- since all the user information is saved in the database, in case of any accident the stored data can be used to find the details of the owner. Online payment system- unlike the traditional offline payment system which provides a printed copy (paper), here we can make the payment online and obtain an e-doc of the details of the payment which is eco friendly as it prevents unnecessary usage of paper.

Disadvantages Cost of implementation is high- using several hardware and software devices increases the overall cost. Replacement of hardware is difficult- once installed, it is difficult to modify the sensors and cameras and also the cost of replacement is also high. Failure of the Cloud- if the cloud database fails the entire mobile application will break down as it’s the core part of the entire application. It may be confusing for unfamiliar users- the users may find it difficult to use the application of follow the instructions specified by the navigation system. Maintenance is high- the hardware needs to be checked frequently to ensure that it is working properly.

and helps to driver/customer to find the empty parking slot in the parking area. The proposed system makes use of a NFC technology (near field communication). In the proposed system the Author has introduced an e-Valet parking service which is similar to the original Valet systems. The main purpose of this system is to have an employee or a third person who takes up the task of finding the empty slots in the park area. Similarly, e-Valet is concept of using electronic means to remove the need of a third person but still functions like valet service [1]. The e-Valet service provides an easy and efficient way for the customer to look and book the available parking slots in the park area. It also enables pre-booking i.e, booking the slots in advance through a mobile based application. Using iSCAPS, the e-Valet mobile application, users can make a request for reserving a slot. Before allocating the slots, the system will look into the database to check the availability of slot. The slots which are vacant can be reserved by the customer. Customers who wish to enter the parking area make use of their smart phones or rent out the NFC cards. The user has to tap the NFC enabled device on a NFC reader which will display the available number of parking slots at each level on the LCD screen. An empty slot will be assigned to the user and the LED at that lot changes from green to blue i.e, it is waiting for the customer to park the vehicle. The same procedure is performed when the customer wants to exit the parking area. Additionally, the amount gets deducted from the NFC device. The system also provides a solution to the problem when the user forgets his/her parked location. To determine the location of the parked vehicle, the user only needs tap the NFC device at an NFC kiosk machine which provides the location of the vehicle as well as the available balance. Table 1: Components and their working

Advantages



More convenient

 

Eco friendly Saves time

Disadvantages

 

Can be costly Complex

[5] In this paper, we propose Vehicular Cyber-Physical Systems (VCPS) along with the cloud support. Here, each layer may provide multiple context-aware services. CLOUD-ASSISTED CONTEXT-AWARE ARCHITECTURE In this section, author has proposed a context-aware architecture with mobile cloud support and two crucial service components. Fig 5 shows a cloud-assisted context-aware architecture. MULTI-LAYERED ARCHITECTURE In the vehicle range, onboard equipment (OBE) installed in the vehicle can provide all kinds of services (e.g., GPS Navigation, Entertainment). In the location computational layer, roadside equipment (RSE) deployed at strategic locations can exchange information with OBE installed on vehicles passing by. Vehicles outside the range of RSE may still be connected to the infrastructure network via neighboring vehicles. This infrastructure network can generate accurate real-time traffic information. In Fig-5, an inter-cloud environment in CVC includes multiple cloud systems running with different policies and different environment with each other to share resources so that end-toend QoS can be maintained even in the event of large fluctuations in computing load which cannot handle by a single cloud system.[9]

Fig 5: example of a Context-Aware architecture WORKING

Author has proposed a framework using Mobile CloudComputing capability and emphasizes cloud-assisted architecture for parking reservation system. Traditional parking garages: The context information of each parking space detected by a Wireless Sensor Networks(WSN), and the internet. The data collected by the sensors are moved to the cloud storages and then selectively transmitted to the users. Also, the status of the parking garages may be dynamically published on nearby billboards to users. Dynamic parking services: In this scenario, we consider a situation in which we may temporarily park a vehicle along the road. For example, during rush hours, there is usually heavy traffic. Therefore, considering this information such as road conditions and rush hours, we may dynamically arrange parking services for a very wide road. With the support of many new technology as Mobile Cloud-Computing (MCC) and Wireless Sensor Networks, the traffic authorities can carry out the dynamic management of this kind of service. In the proposed framework, analysis of three aspects, including service planning of traffic authorities, reservation service process and context-aware optimization is being done. Decision making of traffic authorities: The decision making of the proposed scheme depends on many factors, such as historical traffic flow capacity, road conditions, weather conditions and traffic flow forecasting. In order make effective predictions, researchers need to do data mining and discover useful information and knowledge from collected big data. Parking reservation services: As shown in Fig-6, the status of the parking space can be monitored as shown in the corresponding system, and concurrently updated in the traffic cloud. Through smart phones, users can quickly obtain parking space’s information. Within a given time, we may log into the traffic cloud and subscribe to a parking space.

Figure 6: Context aware dynamic parking services. Context-aware optimization The context information includes not only road conditions and the status of the parking spaces, but also the expected duration of a parking as well. The purpose of the proposed system is to optimize the best locations for drivers. For the parked vehicles, the expected duration of parking can be uploaded to the traffic cloud and the information is shared with the drivers simultaneously. In this way, even when the parking garages has no empty slots, drivers still can inquire and get the desired services by context-aware optimization. The proposed context-aware dynamic parking service offers the solutions to overcome the parking problems faced by the people in their daily life and improving the quality of Contextaware VCPS. Many technologies such as WSN, traffic clouds, data mining and big data are enabling application scenario to become reality. [10] But, due to the shortage of available land per person and city planning being lagging behind, the problem of parking difficulty becomes increasingly severe. To reinforce the management of roadside parking, to call for reasonable roadside parking and to develop a novel roadside parking management system become an effective means to solve the existing problems. Roadside Parking System and Sensor Networks Chronologically, the development of roadside parking management system mainly goes through several stages as follows: 1) Stage of Charging by Manpower: This type of system needs to delimit the parking areas and missionaries are required to charge and find parking space for drivers. Thus, this system has low efficiency. Although it means a low expenditure at first, the cost of manpower is high. 2) Stage of Charging by Parking Meter: Parking meter system is still the basic element of infrastructure of many traffic management systems. It requires drivers to find parking space themselves, but it charges by parking meters. Its advantage is that it leaves out the cost of manpower, but, at the mean time, there are many disadvantages. For example, the drivers have to find parking space blindly without any navigation information that wastes a lot of time. Furthermore, installation of parking meters needs a lot of capital devotion and affects the city's appearance too. In particular, it is hard to solve the problem of high power waste. 3) Stage of ITS: Currently, the Intelligent Transportation System (ITS) has been put into run. ITS requires video cameras to be installed and missionaries watch the picture formed by video camera in control room which is located at remote end and judge the condition of parking space according to the picture. Its disadvantage is that it can't run reliably when the weather is awful, such as rain, fog and snow. Although some systems use radar, sonar or laser radar instead of video camera, which enhance the reliability of the system, the system costs more and has high power waste. This paper presents a sort of system which uses sensor networks to collect

information of parking space. It is an improvement of parking management part of existing ITS infrastructure which can solve the problems mentioned above efficiently. System structure A) Sensor Networks on the street To implement this system, we should deploy sensor networks on the street where parking management is needed. Deployment method can be seen in Fig. 7. It does not need to embed sensors on the road deeply which means low cost. Every sensor detects the presence of vehicle by measuring disturbances in the Earth's magnetic field. In addition, each sensor has luminous device, such as LED, which indicates the reservation information of the parking space.

Fig 7: System Schematics B. Other Parts of the System Small control centre (i.e., base station), which is located at the intersections, is responsible for collecting data from the nearest sensor and judges the state of parking space (i.e., vacancy or occupancy). Each small control centre connects to main control centre through wired network according to mesh network topology and architecture. The parking space state information collected by small control centre will be transmitted to main control centre finally. Main control centre conducts classification process so that users can query it. 4) System Design A. System Overall Design: In this system, there is a sensor at every parking lot and sensor is responsible for detecting the presence of vehicles. Every sensor can communicate with the micro-operating system embedded in the vehicle to identify the information of vehicle. The system detects the presence of vehicles by measuring the disturbance in the Earth's magnetic field. To use power efficiently, microcontroller in each sensor can switch itself to a dormant state. Microcontroller can wake it up periodically so that sensor can collect the information of parking space. As sensor transmits data in a short distance, every sensor communicates with sensors nearby, but several sensors can form sensor networks. The sensor networks in the system degrade to a linear network which is beneficial to avoiding system choke point of data transmission. Every sensor transmits the collected data to small control centre at the intersections by passing through several sensors. In addition, small control centre is responsible for counting users' parking time and transmitting parking time and parking space state information to main control centre.

B. Node structure Fig. 8 describes data flow direction of sensor node, from collecting data to transmitting to neighboring node. Firstly, magnetic sensor detects the presence of vehicle by measuring the disturbances in the Earth's magnetic field. The collected signal is signified and sent to microcontroller and then microcontroller sends the signal to radio transmitter after relevant judgment. Finally, radio transmitter transmits the signal to neighboring node by antenna. The power of the system is supplied by a battery Fig. 9 introduces the integral structure of sensor networks and routing information of parking space state. Several nodes can compose a cluster and data is transmitted from originator to sink node and then to other networks, such as Internet.

configure the network information. The partial simulation of the system is illustrated in Fig. 6:

Through certain measurement. The power waste of functional units in sensor node is listed in the TABLE 2 below. Table 2: Power waste table Fig 8: Sensor Node Structure Schematics

Fig Fig 9: SensorNetwork Structure System Simulation: To make the development of the system more convenient and fast, we use NS2(network simulator 2) to simulate network part of the system. A Brief Introduction to NS2: NS2 is written in C++. For the sake of making it convenient to use, its front end is controlled by OTcl language (objectoriented TCL scripting language). As NS2 supports the simulation of sensor networks, it can simulate the system effectively. In addition, NS2 has a visual windows-based program called Nam (Network Animator) which can show the result of simulation. To use Nam to simulate, we should create a trace file first. In the trace file, there is topological information, such as node location, establishment of data link and time of transmitting data packets etc., which is necessary to simulate the system. NS2 requires writing Tel scripting to

When sensor node is not in an active mode, its power waste is only about 130 ns-A and when in an active mode, its power waste is dependent on the data size transmitted by sensor nodes. Since the data size is small, the power of node is low. III. RESULTS/INFERENCES A. Solutions For The Issues 1. To overcome this problem, we have introduced RFID technology which consists of two main components RFID tag and RFID reader. Firstly, when the user books a slot, the entire details of that information including the user’s information is stored in the database. Once the user enters the Parking Area, the details of the slot location such as level number, and slot number will surely be provided. When the user attempts to park into a slot, the RFID tag which is embedded within the

2.

3.

QR code will be read and if the details are matched then the user can park his/her car. Else the user will be notified about the same. Hence, ensuring that no user can park their vehicle at a wrong slot. To overcome this problem, we have made use of a NFC technology which works as E-Valet and determines the number of vacant slots on each level. Customers who wish to enter the parking area make use of their smart phones or rent out the NFC cards. The user has to tap the NFC enabled device on a NFC reader which will display the available number of parking slots at each level on the LCD screen. We can integrate NFC technology along with Dijkstras algorithm to allocate the closest parking slot hence reducing the amount of time taken to park the vehicle. In case of parking area having multiple entry points, the empty slot which is closest to the customer is determined by using dijkstra algorithm and the slot closest to the entry point is allocated. In order to eliminate the problem of concurrent access by multiple users, we can use scheduling

4.

5.

algorithm such as FCFS (first come first serve). So the customer who initiates the payment first will be allowed to book the slot and the other person will be notified about the same and will be redirected to the next available slot. In such cases when the user only wants to park the vehicle for a specific period of time, the system called parking meter can be used. In this system, once the presence of a vehicle is detected by sensors, the led changes from green to red. As soon as the led changes to red the timer starts and as soon as the vehicle leaves the led changes to green and the timer stops. Once the timer stops, the cost will be calculated accordingly. To overcome this problem, we can make sure that the camera is only turned on when required. This can be achieved by using the data generated by the sensors and as soon as the led turns red i.e, a vehicle is present, the camera will be turned on.

B. Comparison Table Table 3: Comparison of various Methods Parameters

Ultrasonic Sensors

RFID

NFC

Infrared sensors

ITS

RSE

Cost

Low

Moderate

Moderate

Low

High

High

Accuracy

Moderate

High

High

Moderate

High

Moderate

Range

50-70m

100m

Few cm

150cm

Efficiency

Moderate

High

Moderate

Low

Moderate

High

High

High

Low

Moderate

Moderate

Moderate

Maintainability

IV. CONCLUSION The growth of Internet of Things (IoT) and Cloud Computing has led to advancement in parking technology which enables users, the easy and efficient way of booking parking slots. This real-time system provides availability of unreserved parking slots in a particular area. This paper is intended to improve the parking facilities of a city. The use of various sensors such as ultrasonic and infrared sensors allows for tackling and monitoring of data about each slot in real time. Various technologies such as RFID and NFC have to use to provide enhancements in security and reducing the time

-

-

wastage respectively. Also, after the advancements in current Image processing techniques, we can eliminate the use of multiple cameras of detection by using a wide angle camera which when integrated with image processing and neural network techniques can increase the efficiency as well as reduce the overall cost of the system. Also the roadside parking facilities can be improved by the use of techniques such as context aware parking services which makes use various OBEs (On Board Equipment) and RSEs (Road Side Equipment) to improve the overall system performance.

References [1]

[2]

[3]

[4]

[5]

R. Lookmuang, K. Nambut and S. Usanavasin, "Smart parking using IoT technology," 2018 5th International Conference on Business and Industrial Research (ICBIR), Bangkok, 2018, pp. 1-6. L. Kumar, M. H. Khan and M. S. Umar, "Smart parking system using RFID and GSM technology," 2017 International Conference on Multimedia, Signal Processing and Communication Technologies (IMPACT), Aligarh, 2017, pp. 180-184. R. Lookmuang, K. Nambut and S. Usanavasin, "Smart parking using IoT technology," 2018 5th International Conference on Business and Industrial Research (ICBIR), Bangkok, 2018, pp. 1-6 J. T. Ang, S. W. Chin, J. H. Chin, Z. X. Choo and Y. M. Chang, "iSCAPS - Innovative Smart Car Park System integrated with NFC technology and e-Valet function," 2013 World Congress on Computer and Information Technology (WCCIT), Sousse, 2013, pp. 1-6 J. Wan, D. Zhang, S. Zhao, L. T. Yang and J. Lloret, "Context-aware vehicular cyber-physical systems with cloud support: architecture, challenges, and solutions," in IEEE Communications Magazine, vol. 52, no. 8, pp. 106-113, Aug. 2014.M2

[6]

Chen Wenzhi and Liu Bai, "A smart roadside parking navigation system based on sensor networks for ITS," 2006 IET International Conference on Wireless, Mobile and Multimedia Networks, hangzhou, China, 2006, pp. 1-4. [7] S. Shinde, A. Patil, S. Chavan, S. Deshmukh and S. Ingleshwar, "IoT based parking system using Google," 2017 International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC), Palladam, 2017, pp. 634-636. [8] B. M. Mahendra, S. Sonoli, N. Bhat, Raju and T. Raghu, "IoT based sensor enabled smart car parking for advanced driver assistance system," 2017 2nd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT), Bangalore, 2017, pp. 2188-2193. [9] F. Toutain, A. Bouabdallah, R. Zemek and C. Daloz, "Interpersonal context-aware communication services," in IEEE Communications Magazine, vol. 49, no. 1, pp. 68-74, January 2011. [10] S. Al-Sultan, A. H. Al-Bayatti and H. Zedan, "Context-Aware Driver Behavior Detection System in Intelligent Transportation Systems," in IEEE Transactions on Vehicular Technology, vol. 62, no. 9, pp. 42644275,Nov2013.