TRUST-SC: Truthful Multi-Task Double Auction for Quality-Aware Spatial Crowdsourcing in Strategic Environment
Chattu Bhargavi, Vikash Kumar Singh, Alok Kumar Shukla
Spatial crowdsourcing (SC) enables the assignment of location-based tasks to mobile users who must travel to specific locations to perform sensing or service activities. However, SC systems often operate in strategic environments where both task requesters and task executors possess private valuation information, posing challenges for designing efficient and truthful incentive mechanisms. To address these issues, this paper proposes a truthful multi-task double Auction for quality-aware spatial crowdsourcing (TRUST-SC). The proposed framework adopts a three-tier architecture. First, task executors are grouped into spatial clusters to improve scalability and reduce allocation complexity. Second, reliable executors are identified through a majority-voting-based quality evaluation process. Third, tasks are allocated, and payments are determined through a multi-unit double-auction mechanism that guarantees incentive compatibility and individual rationality. Theoretical analysis and simulation results demonstrate that the proposed mechanism achieves efficient task allocation, reliable executor selection, and improved performance compared with existing benchmark mechanisms.
Read on ELI