International Journal of Computer Applications (0975 – 8887) Volume 177 – No. 46, March 2020 30 A Prediction Algorithm for Cloud Integrated Smart Parking System to Search Vacant Parking Spot using Internet of Things Sifat Rahman Ahona Computer Science and Engineering American International University-Bangladesh (AIUB) Dhaka, Bangladesh Naima Hassan Computer Science and Engineering East West University Dhaka, Bangladesh ABSTRACT In recent years it has been observed that the approach of smart city has been increased. It is because of the growing portion of IoT devices for consumers. In order to boost the productiveness and authenticity of metropolitan framework persistent attempts are being made in the field of IoT. Traffic congestion, Road blockage, restricted parking and road safety are some issues that has come forward in IoT. In this paper, a single platform smart parking system integrated to cloud based on IoT is presented. The proposed Smart Parking system comprises certain inventive technologies that can administer car parking automatically. This research also proposes a prediction algorithm to improve efficiency of searching and booking of vacant parking spot and also implements a prototype with Wi-Fi access on a single platform established on Arduino where using a smart phone the system and the vehicle will get an interface to validate the workability of the proposed system. General Terms Internet of things Keywords IOT, Smart parking, wireless, applications, Cloud. 1. INTRODUCTION Everyone who has ever been disappointed driving around metropolitan zones looking for parking has desired a solution that could rapidly lead them to that elusive location. It has been found in a recent research that a driver takes almost 8 minutes to park his vehicle as he invests more time searching the parking garage. This searching leads to thirty to forty percent of traffic congestion [1]. At present, the frequently used method to find parking space is not automated and most of the time user rely on their previous experience. The process of finding parking space gets worst in places where vehicle density is large. One solution is to find a car parking with large capacity. Nonetheless, as the car park might be far from user destination this cannot be considered as an ideal solution. Lately, it has been observed that different technologies of wireless network such as radio frequency identification (RFID), Wireless mesh network has been used to develop interaction between automobile to automobile and automobile to framework. The goal of this research is to give user information about free parking in a nearby place considering road congestion and reserve the space some time earlier through one single platform using their smartphones or other devices. Nevertheless, the present smart parking system available does not give a perfect solution to search a free parking space as this system does not provide solution to some problems such as road congestion, cost effectiveness, and vehicle-denial. In the proposed system every space is considered as an IoT network and also stored the vehicle’s GPS location, length between car parking and the car, quantity of the free parking space will be updated on cloud automatically whenever a parking becomes free or booked. It used a Geolocation Service that calculates the cost of a parking request in terms of distance and time, and this cost is updated periodically and are reachable anytime in the network by the vehicles. The smart parking system is based on certain inventive technologies that is able to administer car parking automatically. It implements a prototype with Wi-Fi access on a single platform established on Arduino using a smart phone that will supply the system and the vehicle an interface to validate the workability. Moreover, in proposed model, each car parking can work individually as a conventional parking system. The smart parking system not only provides benefit to the driver but also beneficial to the managers who manages the parking sports as it maintains a reservation system. Because of the reservation system user can make a reservation earlier before coming to the spot using the proposed app. In contrast to traditional parking system this smart parking system has considerable amount of benefit. Here driver gets the information about the state of the parking space in period manner. Not only that this process also helps the manager of the parking space to utilize the space properly. This paper has suggested an architecture of a smart parking where a prediction algorithm has been introduced. 2. RELATED WORK Fig 1: Architecture and working system [2]