Blockchain and Federated Learning Enabled Smart Traffic Management System for Smart Cities
Vandana Sharma, Tamizharasi Seetharaman, B D Varalakshmi, Amruta Maruti Khangaonkar · 2023
In recent years, traffic management across smart cities has become a global concern due to the increasing number of vehicles and data breaches. Addressing and resolving various security challenges associated with smart cities has become a difficult process as the user connects smart traffic infrastructures in a variety of ways such as smartphones, connected devices, etc. Even a slight compromise in real-time traffic data can lead to adverse consequences as they are highly confidential. Blockchain remains to be an efficient technique that can identify and detect security threats and appropriately manage traffic systems. In this paper, we define a blockchain and federated learning-enabled smart traffic management system for smart cities that effectively manages the real-time traffic status and prevent security vulnerabilities in a decentralized manner. The use of the federated learning technique improves the system's efficiency through the use of local and global models. It is observed from the experimental results that the proposed approach remains resistant to byzantine attack and Sybil attacks providing more resilient services to the end users.