Research on Spatiotemporal Evolution Methods Networks Based on Deep Learning: Arctic Shipping Data as a Case

ABSTRACT This article uses deep learning to study the spatiotemporal evolution method of vehicular ad‐hoc networks (VANETs). Firstly, the research background and significance are introduced, as well as the research status of VANET spatiotemporal evolution methods and deep learning‐based VANET spatiotemporal evolution methods both domestically and internationally. Then, the related theoretical foundations of deep learning and traditional VANET spatiotemporal evolution methods are elaborated in detail. Subsequently, three VANET spatiotemporal evolution methods based on long short‐term memory‐convolutional neural network (LSTM‐CNN), deep time generative adversarial network (D‐TGAN), and fully connected network (FCN) are proposed, and simulation experiments are conducted for each method to analyze the experimental results. Finally, the main work of the whole article is summarized, and future research directions are discussed. Through this study, new ideas and methods can be provided for the development of VANET spatiotemporal evolution methods.