Privacy-Aware Multi-task Allocation for Hybrid Blockchain-enabled Mobile Crowdsensing with Wireless Sensor Networks

Mobile crowdsensing with wireless sensor networks (MCS-WSN) emerges as a promising sensing paradigm to collect large-scale sensing data through WSN in a cost-effective manner by outsourcing the MCS tasks to mobile users. However, the sensing data contributed by the mobile participants usually contain users' private information, which raises considerable concerns about privacy and trust issues. On the other hand, with the accumulation of sensing data, the large computing overhead of data aggregation in WSN becomes a non-negligible factor affecting the system performance. To address the above issue, in this paper, we introduce a hybrid-blockchain-enabled MCS-WSN platform. The hybrid blockchain is adopted to assist the MCS task release and data validation. Meanwhile, the cloud-edge architecture is also applied in WSN to enable reliable data transmission and data aggregation. The multi-task allocation and computing offloading decision are modeled as a Markov Decision Process (MDP) and solved through deep reinforcement learning (DRL), considering both platform utility and task execution latency. The simulation results demonstrate the efficiency of the proposed approach.