Abstract:
Spatial crowdsourcing, e.g., Vehicular Ad-Hoc Network (VANET)-based spatial crowdsourcing, is a new distributed computing paradigm, in which task assignments highly rely ...Show MoreMetadata
Abstract:
Spatial crowdsourcing, e.g., Vehicular Ad-Hoc Network (VANET)-based spatial crowdsourcing, is a new distributed computing paradigm, in which task assignments highly rely on the wisdom of the crowdsourcing platform. However, with the increase in user data leakage incidents, the existing task assignment methods are no longer sufficient to meet the privacy requirements of users. In the existing VANET-based spatial crowdsourcing, task assignments are usually performed by a trusted third party based on the real locations of tasks and drivers (task performers), which may lead to the leakage of the users' locations. Furthermore, the drivers usually prefer to query the nearest tasks to them in a geometric range, at this point sending query requests to remote crowdsourcing servers increases unnecessary response delays. To assign tasks securely and efficiently, we propose a privacy-preserving task assignment scheme based on OT and edge computing (PriTAEC), which is the first to apply Oblivious Transfer (OT) and edge computing to preserve the location privacy of VANET-based spatial crowdsourcing. In the scheme, we first utilize Hilbert Curve and Bloom Filter to implement location range queries. Then, we use geohash location encoding and Oblivious Transfer to achieve fine-grained location matching. In particular, we design a task assignment algorithm with an offline-online phase to improve the efficiency of task assignments. Finally, we prove the security of the scheme and evaluate its performance, which shows our scheme is secure and efficient.
Published in: IEEE Transactions on Vehicular Technology ( Volume: 72, Issue: 4, April 2023)