GPU-Accelerated Compression and Visualization of Large-Scale Vessel Trajectories in Maritime IoT Industries

Yu Huang,Yan Li,Zhaofeng Zhang,Ryan Wen Liu
DOI: https://doi.org/10.1109/jiot.2020.2989398
IF: 10.6
2020-11-01
IEEE Internet of Things Journal
Abstract:The automatic identification system (AIS), an automatic vessel-tracking system, has been widely adopted to perform intelligent traffic management and collision avoidance services in maritime Internet-of-Things (IoT) industries. With the rapid development of maritime transportation, tremendous numbers of AIS-based vessel trajectory data have been collected, which make trajectory data compression imperative and challenging. This article mainly focuses on the compression and visualization of large-scale vessel trajectories and their graphics processing unit (GPU)-accelerated implementations. The visualization was implemented to investigate the influence of compression on vessel trajectory data quality. In particular, the Douglas–Peucker (DP) and kernel density estimation (KDE) algorithms, respectively, utilized for trajectory compression and visualization, were significantly accelerated through the massively parallel computation capabilities of the GPU architecture. Comprehensive experiments on trajectory compression and visualization have been conducted on large-scale AIS data of recording ship movements collected from three different water areas, i.e., the South Channel of Yangtze River Estuary, the Chengshan Jiao Promontory, and the Zhoushan Islands. Experimental results illustrated that: 1) the proposed GPU-based parallel implementation frameworks could significantly reduce the computational time for both trajectory compression and visualization; 2) the influence of compressed vessel trajectories on trajectory visualization could be negligible if the compression threshold was selected suitably; and 3) the Gaussian kernel was capable of generating more appropriate KDE-based visualization performance by comparing with other seven kernel functions.
computer science, information systems,telecommunications,engineering, electrical & electronic
What problem does this paper attempt to address?
The paper attempts to address the problem of efficiently compressing and visualizing large-scale vessel trajectory data in the maritime Internet of Things (IoT) industry. With the rapid development of maritime transportation, the volume of vessel trajectory data based on the Automatic Identification System (AIS) has increased dramatically, making trajectory data compression both important and challenging. The paper mainly focuses on the compression and visualization of large-scale vessel trajectories and their accelerated implementation on the Graphics Processing Unit (GPU). Specifically, the paper utilizes the Douglas-Peucker (DP) algorithm for trajectory compression and significantly accelerates this process through the powerful parallel computing capabilities of the GPU. Additionally, to evaluate the impact of compression on the quality of vessel trajectory data, trajectory visualization is also implemented. During the visualization process, the Kernel Density Estimation (KDE) algorithm is used to generate density maps, which are similarly accelerated by the GPU to improve efficiency. The main contributions of the paper include: 1. Redesigning the hierarchical structure of the DP-based trajectory compression algorithm according to the GPU architecture and programming framework, which significantly speeds up the compression of large-scale vessel trajectories while maintaining compression quality. 2. Proposing a GPU-based trajectory visualization framework that mainly includes parallel data projection and interpolation, as well as reducing memory communication to accelerate KDE convolution, ensuring a significant reduction in computation time under different experimental conditions while maintaining visualization performance. 3. Experimental results show that the proposed GPU-based parallel framework outperforms the standard CPU computation baseline in both trajectory compression and visualization, particularly excelling in large-scale vessel trajectories collected from three different waters. Overall, this study greatly improves the efficiency of trajectory compression and visualization by fully leveraging the massive parallel computing capabilities of the GPU, which is of great significance for enhancing the safety and efficiency of maritime applications.