A Smart and Secure IoT-Enabled Parking System with ML Integration

Publications

A Smart and Secure IoT-Enabled Parking System with ML Integration

Year : 2026

Publisher : Elsevier B.V.

Source Title : Procedia Computer Science

Document Type :

Abstract

In this modern era, a Smart Parking System (SPS) is necessary for maintaining the traffic flow. Smooth traffic flow reduces road congestion and improves the overall parking experience for the public. The existing system faces several challenges, including unproductive space utilization and dependence on manual operations. Manual operations are often vulnerable to human error. To overcome these limitations, this paper proposes a comprehensive IoT-enabled SPS that integrates Hyperledger Fabric (HF), the permissioned blockchain framework. The real-time vehicle detection by ultrasonic sensors, environmental monitoring sensors, and wireless communication for efficient data transmission are integrated into the SPS. The proposed model is implemented by using a sensor and communication module-based hardware simulation. Blockchain technology (BCT) integration ensures the immutability of parking records present inside the distributed ledger. Moreover, BCT also provides integrity and transparency of the data. A miniature implementation of ML is used to predict the basic things, such as parking occupancy and sensor status, which adds flexibility to the system. Additionally, certain performance parameters are evaluated for the SPS, such as success rate, average transaction creation time, and latency. Moreover, a web-based dashboard is constructed for monitoring real-time data like parking availability, blockchain statistics, and sensor status. Simulation results depict the successful integration of IoT and BCT, with an average creation time of 15008 ms, an average success rate of 99.4%, and an average reading latency of 13631 ms. Thus, it ensures the reliability and secure management of parking data. Furthermore, simulations prove that the SPS is performing robustly in differing network conditions.