Deadline-constrained Data Aggregation Scheduling in Urban Vehicular Networks

With the rapid increasing scale of Intelligent Transportation System (ITS) and vehicular network, our quality of life are significantly improved, especially traveling in urban area. To supply a large amount of real-time ITS applications, such as navigation and automatic driving, aggregation delay should be controlled in a small range so that more data might be dropped during aggregation. State-of-the-art data aggregation approaches determine to aggregate and forward data based on the instantaneous link status. However, wireless link status in a period should be taken into account to reduce dropped data in a long term. Given the estimated further link status, this paper formulates deadline-constrained data aggregation scheduling problem into a nonlinear integer programming problem. We design an efficient algorithm, DASV, to accelerate branching the solution space because most of the variables appear only once in the objective function or constraints. Based on taxi trajectory data in San Francisco, experimental results demonstrate the proposed DASV outperforms state-of-art methods 22.7 % in terms of aggregation ratio.