Open Access

ARTICLE

Privacy-Preserving Collaborative Task Allocation for Multi-Skill Mobile Crowdsensing

Jie Li, Fuyuan Song^*, Qin Jiang

School of Computer Science, Nanjing University of Information Science and Technology, Nanjing, China

* Corresponding Author: Fuyuan Song. Email:

Computers, Materials & Continua 2026, 89(1), 95 https://doi.org/10.32604/cmc.2026.085946

Received 21 May 2026; Accepted 17 July 2026; Issue published 13 August 2026

Abstract

Mobile crowdsensing enables large-scale sensing tasks through smart devices carried by users and has been widely applied in intelligent transportation and environmental monitoring. With the increasing complexity of sensing tasks, many tasks require the collaboration of multiple workers with different skills. However, both task-required skills and worker skills are privacy-sensitive, and directly exposing them to the platform may reveal task intentions and workers’ capability profiles. To address this issue, this paper proposes Dual-Fog Privacy-Preserving Multi-skill Task Allocation (DPMTA), a privacy-preserving task allocation scheme for multi-skill collaborative tasks. DPMTA adopts a dual-fog architecture to separately protect location privacy and skill privacy. Specifically, task and worker locations are perturbed by differential privacy, while skill information is split by XOR secret sharing and distributed to two non-colluding fog servers. With Beaver triple-assisted secure Boolean computation, DPMTA enables skill matching, coverage updating, and collaborative worker selection without revealing plaintext skills. In addition, a random permutation mechanism is introduced to hide the direct mapping between skill bit positions and semantic skill labels. Security analysis shows that DPMTA effectively protects skill and location privacy. Meanwhile, the experimental evaluation demonstrates that DPMTA maintains a high task allocation success rate while keeping communication and computation overhead within acceptable limits.

---

Keywords

---