Answers are grounded in this paper and cite their sources.
In modern Intelligent Transportation Systems, connected vehicles continuously exchange sensory data and model updates, exposing the network to privacy risks ranging from location tracking to data inference and model inversion. Existing surveys rarely map privacy targets, techniques, application scenarios, and system architectures together under the latency and utility constraints of the Internet of Vehicles. This systematic review is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement and synthesizes 130 primary journal articles published between 2021 and 2026. The methodology combines content coding with BERTopic modeling, where topic membership is significantly associated with manual codes. Technical countermeasures concentrate on three main categories, namely pseudonyms for communication, Differential Privacy for statistical release, and Homomorphic Encryption for ciphertext computation. By evaluating these mechanisms across Vehicle to Everything messaging, Federated Learning, and data sharing, this review demonstrates how privacy can be protected while maintaining data utility.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: