Abstract: |
The interface defined in this paper provides a generic way to select, structure, aggregate, and retrieve spatially and/or temporally localized measurement data from an underlying database in a standardized manner. It is generic in the sense that it is neither specific to a particular type of database, nor is it specific to a particular programming language or data format. The interface is particularly designed for software systems where statistical analyses of potentially large collections of scalar measurement data are to be performed by loosely coupled client applications. A key feature of the interface design is that a statistical aggregation and evaluation of the data is performed on the server side, such that the necessary amount of data to be transferred to a querying client is minimized. This can be a crucial feature for clients that are querying large databases remotely, for example, over the Internet. The interface delivers regularly arranged data to facilitate statistical assessments (e.g., visualizations in charts). It does not require, however, that the raw data is arranged regularly in any way. In particular, measurements are not required to be synchronous or equally spaced. The proposed interface can be employed for a wide range of application areas, e.g., to evaluate data from sensor networks measuring the street traffic, the water and energy supply, the air pollution or climate indicators. |