Distributed Downloading Strategy for Multi-Source Data Fusion in Edge-Enabled Vehicular Network : (Invited Paper)
Yue Yu, Jun Yong Wu, Xiao Tang, Tiecheng Song, BaekGyu Kim, Zhu Han · 2019
Multi-source data fusion to support intelligent transportation system (ITS) is a promising service offered by mobile edge computing (MEC). With the fusion results delivered in near real-time, drivers or autonomous vehicles can peak around the corner, extend sensing range, reinforce and validate local observations to make safer and smarter driving decisions. However, downloading too much data increases the service delay thus undermines the fusion computing service performance. In this paper, we analyze the optimal downloading strategies of vehicles. By establishing the optimization indicator to monitor and evaluate fusion computing service, we use a hierarchical game, which is equivalent to a mathematical programming with equilibrium constraints (MPEC), to formulate the intersection between the MEC and vehicles. Through analysis, we transform the MPEC problem into a solvable single-layer optimization problem. We also provide an unpractical centralized approach, which has immense signaling overhead and exponentially-growing complexity, as a performance upper bound. Numerical results validate the theoretical analysis and demonstrate that the proposed downloading strategy has near-optimal performance in terms of system utility and service delay.