Preserving Location Privacy and Accurate Task Allocation in Edge-assisted Mobile Crowdsensing

Abstract

Mobile crowdsensing enables collaborative data sensing between cloud server and mobile nodes. To participate in the sensing task, mobile nodes upload their locations to the centralized cloud for task allocation. However, revealing locations to an untrusted cloud results in privacy leakage, such as trajectories tracking and home address exposal, threatening the personal security. Obfuscation and cryptography based schemes are two main solutions to protect the location privacy. However, these schemes may either degrade the accuracy of task allocation or rely on some strong assumptions. Thus, how to protect location privacy without strong assumptions while remaining high accuracy in task allocation is challenging. In this paper, we propose a secure protocol for edge-assisted mobile crowdsensing, which removes the assumption that the cloud cannot collude with mobile nodes. Specifically, we deploy homomorphic encryption among service requestor, cloud server and edge nodes in a collaborative manner. Benefiting from the additive property of the cryptosystem, the cloud is able to securely calculate the mobile node’s travel distance while knowing nothing about the mobile mode’s location and task location. Based on the protocol, two types of location-dependent task allocation, travel distance based task allocation and spatial distribution based task allocation, can be implemented with location privacy preservation. Experimental results show the effectiveness of our work in task allocation. In addition, comprehensive privacy discussion indicates that the proposed protocol is secure from the collusion between cloud and mobile nodes, while preserving the task location and location privacy of mobile nodes.