Real time transducer signal features extraction: A standard approach

This paper presents a real time algorithm, based on a segmentation and labelling technique, for analog transducer signals, which extracts information from the signal. The process of sampling is extremely important because it is the only one that connects the real world with the digital world. In addition, the proposed technique can be embedded into sensors, allowing applications in sensor networks and sensors data fusion architectures like Internet of Things. A simplified version of these algorithms has been proposed as a standard for transducer signal feature extraction. These algorithms are analysed and experimental results are tackled for representative signals, in different fields.