>> 《Geospatial Information Technology for Emergency Response》简述

Part 3 Data management and routing in 3D

---

Real time and spatiotemporal data indexing for sensor based databases　

S. Servigne & G. Noel
LIRIS: Research Center for Images and Information Systems, INSA, Villeurbanne Cedex, France

ABSTRACT: The most important aspect of a system for emergency response is time. So it is necessary to organise fast storage of newly arriving data into databases, fast search of data, maintenance of time sequences, robustness of the systems. All of these processes have to be real time. Another important aspect for environmental risk monitoring for example, is to allow real-time queries based on spatiotemporal attributes. Fastest accesses to the most recent data are also necessary. In addition, real-time data collected into real time database from sensors require fast and efficient management to avoid main memory saturation. The chapter discusses real-time spatiotemporal indexing schemes and real time management of main memory. Some possible solutions for real time spatiotemporal data indexing are presented to allow real time and fast data structuring. The proposed index methods support queries privileging recent data. One of the presented indexing solutions is dedicated to real time spatiotemporal data issued from fixed sensors. Another one is dedicated to real time spatiotemporal data collected from agile sensors. Finally, an outline solution concerning management of real time memory saturation according to the significance of data is presented.