In this paper, we investigate the feasibility of a hybrid radio/accelerometric approach to perform arm posture recognition. A radio fingerprinting-based approach, through measurements of the Received radio Signal Strengths (RSSs) from anchor nodes, is first used to localize the positions (among a set determined during a training phase) of target nodes properly placed on a user arm. Accelerometric signals generated by the target nodes are then used to estimate the pitch of every device in order to refine the radio fingerprinting results and perform posture recognition, i.e., “continuous” estimation of the positions of the target nodes. We experimentally investigate, through a SunSPOT wireless sensor network testbed, different fingerprinting-based localization algorithms, namely deterministic and probabilistic. In each case, the system parameters are optimized by minimizing a properly defined Position Error (PE). Finally, a comparison between the performance of the proposed system and that of a low-cost optical arm posture recognition system (namely, Kinect) is presented.

A hybrid radio/accelerometric approach to arm posture recognition

MARTALO', MARCO;
2015-01-01

Abstract

In this paper, we investigate the feasibility of a hybrid radio/accelerometric approach to perform arm posture recognition. A radio fingerprinting-based approach, through measurements of the Received radio Signal Strengths (RSSs) from anchor nodes, is first used to localize the positions (among a set determined during a training phase) of target nodes properly placed on a user arm. Accelerometric signals generated by the target nodes are then used to estimate the pitch of every device in order to refine the radio fingerprinting results and perform posture recognition, i.e., “continuous” estimation of the positions of the target nodes. We experimentally investigate, through a SunSPOT wireless sensor network testbed, different fingerprinting-based localization algorithms, namely deterministic and probabilistic. In each case, the system parameters are optimized by minimizing a properly defined Position Error (PE). Finally, a comparison between the performance of the proposed system and that of a low-cost optical arm posture recognition system (namely, Kinect) is presented.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11389/20958
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact