An adaptable architecture for river quality monitoring

摘要

This paper describes the architectural approach for Environmental Monitoring Systems and its application to the area of quality monitoring of river water. It presents the first emerging results of the ESPRIT project EMS~1. Based ont eh needs for water quality monitoring the requirements for an architecture capable to link distributed monitoring stations and laboratory databases are identified. Especially for adaptable and upgradable systems, information sources have to be described on a more abstract level. For this purpose a Logical Sensor Model approach has been adopted allowign the specification of data and control flow on different abstraction levels. Special emphasis is put on the three fusion layers: Sensor Validation, Situation Description, and Situation Assessment. Flexible and adaptable systems must be able to make their own decisions during a fusion process such as selecting specific sensors or applying different fusion techniques. In order to do this, an architecture concept based on Distributed AI principles is proposed.