DPDT: A Differentially Private Crowd-Sensed Data Trading Mechanism

摘要: Along with the generation of Internet Things (IoT), values tremendous volumes sensing data will be slowly unlocked. Thus, crowd-sensed trading as a new business paradigm has recently attracted increasing attention. A typical system contains platform, consumers, and crowd workers. The platform recruits workers to collect then sells consumers. In this article, we design differentially private mechanism, called DPDT, preserve identity privacy consumers task against during collection process, simultaneously. DPDT consists auction-based pricing algorithm algorithm. achieves good approximation maximum revenue. Meanwhile, it guarantees $(e^{2}-1)\epsilon $ -truthfulness $2\epsilon -differential privacy, where $\epsilon >0$ is small constant. able effectively protect We prove that $\delta -approximate constant, meanwhile tight bound expected ratio. At last, extensive simulations are conducted verify significant performance DPDT.