Urban Sensing of Virtual Internet of Things for Metaverse

Urban sensing encourages the incredibly quickadvancement of digitalization, intelligence, and ubiquitousperception, accelerating the construction of the metaverse.Various sensors construct the sensing network to capturehuman activities for the human-centered metaverse. How-ever, effective management of ubiquitous urban sensing withvarious attributes is a challenge. In this article, we pro-pose a virtual Internet of Things (VIoT) enabled by digitaltwin (DT) for building the metaverse. Specifically, we intro-duce an architecture of the VIoT, including a policy sensinglayer, a 6G edge-cloud collaboration layer, a DT layer, anda user-friendly terminal layer. Furthermore, to address thesensing scheduling issue in the VIoT, we formulate a sens-ing profit maximization problem by considering the sensingcoverage, data utility, and energy cost attributes of visualsensors and fabric sensors. To tackle this problem effi-ciently, we design a sensing scheduling policy based on thesoft actor-critic (SSP-SAC) algorithm. The simulation resultsdemonstrate that compared to the baseline schemes, theSSP-SAC scheme can significantly improve the sensing profitin diverse situations, indicating that the VIoT can provide aneffective urban sensing policy