Inductive as a support of deductive data visualisation in wireless sensor networks

Wireless Sensor Networks (WSN) have been useful in a variety of domains. These types of networks have an intimate interaction, via sensors, with the physical environment they operate within. The part of the world with which an application is concerned is defined as that application's domain. This paper advocates that the application domain can serve as a supplement to sense data analysis, interpretation, and visualisation methods and tools. To achieve this, we propose a Multi-Dimensional Application Domain-driven (M-DAD) information extraction and visualisation framework that uses the application domain to accurately visualise multimodal sense data. M-DAD harnesses the inherent redundancies and relationships among the collected sense data as information about a specific event of interest in a WSN is usually captured in multiple sensed modalities. The proposed mapping framework utilises these correlations, defined in the application domain, to visualise a sense modality, e.g, soil nitrate levels, using other related but independent sense modalities, e.g, temperature and pH, which results in higher accuracy visualisations than visualising from a single sense modality. The primary experimental results demonstrate that the proposed framework performs as well or better than methods without its extended capabilities.