OCT: A Novel Opportunistic Compression and Transmission Approach for Private Car Trajectory Data

Abstract

The advances in mobile sensing and vehicle cloud service techniques have generated massive spatial-temporal trajectory data, which has caused the crises of storage and communication. In this paper, we propose a novel Opportunistic Compression and Transmission approach, namely OCT, with aims of reducing trajectory transmission overhead and storage cost. We first present the design of trajectory collection terminal based on GPS & OBD wherein the main process of OCT can be implemented rather than the cloud server. Within the proposed OCT, we devise a map-matching method based on MIV-matching and calibrating trajectory, which significantly reduces sampling errors of raw trajectories. Then we divide trajectory data into two parts, i.e. spatial and temporal parts, and realize compression operation separately. By using a prediction model based on historical trajectory velocity, we make use of opportunistic transmission of trajectory data from the GPS & OBD terminal to vehicle cloud server and thus dramatically decrease the transmission overhead. The proposed OCT not only realizes real-time trajectory preprocessing and compressing, but also ensures high trajectory compression ratio. To validate the performance of the OCT, we collect a large-scale private car trajectory data from real urban environments. Extensive experiments verify the effectiveness and superiority of the proposed method.