TMVcrowd: An Authorized and Fine-Grained Encrypted Task Matching Framework on Blockchain for Vehicular Crowdsourcing

Vehicular crowdsourcing has emerged as a promising paradigm that leverages the sensing and computational capabilities of connected vehicles to perform large-scale data collection and task execution. However, existing vehicular crowdsourcing platforms typically rely on centralized servers for task publishing and matching, which introduces risks of single points of failure, privacy leakage, and limited scalability. Blockchain offers decentralization and transparency but poses new challenges, as directly outsourcing sensitive task information onto a public ledger may lead to severe privacy violations. To address these challenges, we present TMVcrowd, a blockchain-based framework for authorized and fine-grained encrypted task matching. TMVcrowd enables secure collaboration between task requesters and vehicular workers by integrating attribute-based access control and efficient searchable encryption. Specifically, we design a novel constant-length ciphertext attribute-based encryption (CL-ABE) scheme to minimize on-chain storage costs and propose a hybrid encrypted task matching method supporting both keyword and range queries. We formally prove the security of our scheme and implement TMVcrowd on Ethereum. Experimental results demonstrate that TMVcrowd achieves strong privacy guarantees, significant storage efficiency, and practical performance, making it suitable for large-scale vehicular crowdsourcing systems.