ADDING A VIRTUAL LAYER IN A SENSOR NETWORK TO IMPROVE MEASUREMENT RELIABILITY

A layer of soft sensors based on neural network is designed and trained at the aim to constitute a virtual layer of measure in a sensor network. Each soft sensor of the layer esteems the missing values of some hardware sensors by using the values obtained from some other sensors performing a spatial forecasting. The correlation analysis for each parameter taken into account is used to define cluster of real sensors used as sources of measure to esteem missing values. An application concerning the fire prevention field is used as test case and result evaluation.