Litcius/Paper detail

Connected Vehicles Secure Data Sharing using Secure and Differential Privacy computation on multi-party

Kamal Kumar, Rallabandi Venkata Santoshi Saraswati Swetha Nagini, A. Shivaprasad, R. Maheswari, Haider Alabdeli, Dilli Ganesh

202518 citationsDOI

Abstract

To ensure effective and secure transportation systems, networked automobiles need to share enormous amounts of data that safeguard users' privacy. Data leakage issues, large computing expense, and lack of differential privacy guarantees are some of the challenges with existing methods. The Privacy-Preserving Vehicular Data Exchange Framework (PPVDEF) to address the above issues. It integrates differential privacy (DP) with secure multi-party computation (SMPC) to render data exchange among infrastructure, cloud, and automobiles safe and speedy. While PPVDEF supports collective intelligence through privacy-conscious data analytics and aggregation, it ensures sensitive vehicle data remains confidential. The proposed methodology enhances security through unwanted access blocking and ensures privacy by adding controlled noise to keep personal data private. The study indicates that PPVDEF makes vehicle data analytics significantly more accurate, efficient, and scalable without compromising data privacy. The approach encourages smarter and safer transportation networks by providing a platform where connected cars can trust each other and their privacy can be ensured. The proposed Method achieves the data privacy by 97.6%, accuracy by 98.5% and scalability by 96.3%.

Topics & Concepts

Computer scienceDifferential privacySecure multi-party computationComputer securitySecure two-party computationInternet privacyDifferential (mechanical device)ComputationData sharingCryptographyComputer networkData miningAlgorithmEngineeringMedicinePathologyAlternative medicineAerospace engineeringPrivacy-Preserving Technologies in DataCryptography and Data SecurityVehicular Ad Hoc Networks (VANETs)
Connected Vehicles Secure Data Sharing using Secure and Differential Privacy computation on multi-party | Litcius