The recent years have witnessed the rise of an enormous number of software algorithms that implement pedometers (or step counters), which led to the development of several context-aware IoT-based smartphone apps for sports and healthcare, among others. While the number of scientific works in this context is high, there is no comparison study that analyzes the different proposal at implementation level. In this paper we first perform a literature review of software implementations of pedometers for smartphones and then classify them into a taxonomy. With this, we highlight the similarities of their scheme, which is based on a number of defined steps to be applied in a pipeline. We then develop a smartphone application that implements all the configurations of these steps found in literature and evaluates them in various scenarios. Finally, we present comparative results obtained by running extensive and real tests that show the importance of a carefully designed filtering step.
Pedometers for Smartphones: Analysis and Comparison of Real-Time Algorithms / Neri, Giacomo; Montori, Federico; Gigli, Lorenzo; Bedogni, Luca; Di Felice, Marco; Bononi, Luciano. - ELETTRONICO. - (2022), pp. 1-6. (Intervento presentato al convegno 2022 IEEE 8th World Forum on Internet of Things (WF-IoT) tenutosi a Yokohama, Japan nel 26 October 2022 - 11 November) [10.1109/wf-iot54382.2022.10152104].
Pedometers for Smartphones: Analysis and Comparison of Real-Time Algorithms
Neri, GiacomoCo-primo
;Montori, Federico
Co-primo
;Gigli, Lorenzo;Bedogni, Luca;Di Felice, Marco;Bononi, LucianoUltimo
2022
Abstract
The recent years have witnessed the rise of an enormous number of software algorithms that implement pedometers (or step counters), which led to the development of several context-aware IoT-based smartphone apps for sports and healthcare, among others. While the number of scientific works in this context is high, there is no comparison study that analyzes the different proposal at implementation level. In this paper we first perform a literature review of software implementations of pedometers for smartphones and then classify them into a taxonomy. With this, we highlight the similarities of their scheme, which is based on a number of defined steps to be applied in a pipeline. We then develop a smartphone application that implements all the configurations of these steps found in literature and evaluates them in various scenarios. Finally, we present comparative results obtained by running extensive and real tests that show the importance of a carefully designed filtering step.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.