TDengine-S2-LSH:A High-Performance Framework for Inland AIS Trajectory Storage and Query

This paper addresses the challenges in managing and querying large-scale AIS (Automatic Identification System) spatiotemporal trajectory data from inland waterways. Due to the complex nature of inland navigation environments, traditional database approaches struggle with the volume, velocity, and complexity of AIS data. The research proposes a comprehensive query service system using TDengine time-series database and advanced spatiotemporal indexing techniques. The system incorporates data preprocessing, efficient storage models, and novel query methods. Implementation using Django framework and distributed computing demonstrates high performance, with query accuracy reaching 100% for basic queries and exceeding 95% for similarity queries, with response times under 5 seconds. The system provides effective visualization and analytical capabilities for maritime management and research.