Astute Video Transmission for Geographically Dispersed Devices in Visual IoT Systems

Wen Ji,Lingyu Duan,Xi Huang,Yueting Chai
DOI: https://doi.org/10.1109/tmc.2020.3009745
IF: 6.075
2020-01-01
IEEE Transactions on Mobile Computing
Abstract:Visual IoT (VIoT) is a promising IoT paradigm that visualizes sensing data from massive numbers of dispersed devices. A key objective in VIoT is to efficiently manage the devices to perform complex task-related visual data processing. Prior multimedia IoT systems have mainly focused on the delivery of captured video to remote servers, without considering the video tasks' characteristics and the devices' heterogeneous capabilities. In this work, we propose an astute video transmission framework for such a VIoT system composed of heterogeneous visual devices. First, we formulate the problem of joint video task allocation and heterogeneous device management by constructing a device hypergraph (DH) structure, which enables devices with different capabilities to perform complex video tasks cooperatively. Second, we model the video transmission within a VIoT system by applying fractal theory considering the NP-hardness of the optimization. In particular, we construct a comprehensive fractal submodular optimization framework through a DH and explore the inner submodular property to effectively leverage both video-task complexity and device heterogeneity. Third, we consider the geographically dispersed characteristic of massive numbers of VIoT devices and propose a multi-hop dispersed transmission mechanism for achieving globally cooperative optimality. The proposed architecture has been evaluated under diverse parameter settings. Numerical results are provided to validate the proposed algorithm in terms of delay, computational efficiency, and bandwidth utilization. Simulation results confirm the effectiveness and superiority of the proposed method.
What problem does this paper attempt to address?