A distributed soft sensors model for managing vague and uncertain multimedia communications using information fusion techniques

Abstract

In wireless sensor networks it may occur that values cannot be measured directly by sensors or are costly. In those scenarios, the use of soft sensors is a promising solution, particularly when dealing with multimodal sensors that record multimedia information. In this paper we propose an approach in which soft sensors work by fusing the measured ozone level in outdoor air recorded by nearby hard sensors. We used a fusion technique based on the idea of favorite peer selection. The statistical indicators obtained indicate that our approach overcomes the baseline in 100% of cases analyzed in 3 different test scenarios.

Keywords

• Soft Sensors
• Information Fusion
• Information Quality
• Sustainability