Information Fusion Using Ontology-Based Communication between Agents

The distribution of on-line applications among network nodes may require obtaining acceptable results from data analysis of multiple sensors. Such sensors data is probably heterogeneous, inconsistent, and of different types. Therefore, multiple sensor data fusion is required. Here, there are many levels of information fusion (from low level signals to high level knowledge). Agents for monitoring application field events could be used to dynamically react to those events and to take appropriate actions. In a dynamic environment even a single agent may have varying capabilities to sense that environment. The situation becomes more complex when various heterogeneous agents need to communicate with each other. Ontologies offer significant benefits to multi-agent systems. The benefits as such are interoperability, reusability, support for multi-agent systems development activities such as system analysis and agent knowledge modeling. Ontologies support multi-agent systems operations such as agent communication and reasoning. The proposed agent based model in this paper can afford a promising model for obtaining acceptable information in case of multiple sensors.