Evaluation of a Dynamic Tree Structure for Indexing Query Regions on Streaming Geospatial Data

Most recent research on querying and managing data streams has concentrated on traditional data models where the data come in the form of tuples or XML data. Complex types of streaming data, in particular spatio-temporal data, have primarily been investigated in the context of moving objects and location-aware services. In this paper, we study query processing and optimization aspects for streaming RemotelySensed Imagery (RSI) data. Streaming RSI is typical for the vast amount of imaging satellites orbiting the Earth, and it exhibits certain characteristics that make it very attractive to tailored query optimization techniques. Our approach uses a Dynamic Cascade Tree (DCT) to (1) index spatio-temporal query regions associated with continuous user queries and (2) efficiently determine what incoming RSI data is relevant to what queries. The DCT supports the processing of different types of RSI data, ranging from point data to more general spatial extents in which the incoming imagery c...

• Type: Conference
• Year: 2005
• Where: SSD
• Authors: Quinn Hart, Michael Gertz, Jie Zhang
• Keywords: Particular Spatio-temporal Data, RSI Data, SSD 2005, Traditional Data Models