A Self-organizing Approach to Activity Recognition with Wireless Sensors

Holzmann, Clemens; Haslgrübler, Michael

doi:10.1007/978-3-642-10865-5_20

Cited by 2 publications

(1 citation statement)

References 8 publications

(9 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various projects have been carried out to build such systems. Holzmann and Haslgrübler (2009) and Kurz and Ferscha (2010) proposed an approach to individual activity recognition which based on a self-organizing of body-worn sensor which enable dynamic sensors compositions according to a recognition goal. Similar approaches proposed by Bao and Intille (2004), Lee and Mase (2002) and Parkka et al (2006) that aim to develop wearable systems that can be used to recognize individual activities.…”

Section: Introductionmentioning

confidence: 99%

Situation‐awareness and reasoning using uncertain context in mobile peer‐to‐peer environments

Wibisono

Zaslavsky

Ling

2013

International Journal of Pervasive Computing and Communications

View full text Add to dashboard Cite

Purpose -The recent advances of mobile computing and sensing technologies have enabled mobile devices to individually sense environment context and develop situation awareness capability. To gain a better understanding of the environment, mobile devices that are co-located can establish a mobile peer-to-peer (MP2P) environment to share their individual context information. The purpose of this paper is to propose a theoretical model for representing and reasoning about situations using uncertain context information captured by multiple devices in an MP2P environment. Design/methodology/approach -The paper proposes a generic model for reasoning about situations using uncertain context information captured by multiple devices in a MP2P environment. A data fusion technique is then integrated into the proposed model. To deal with uncertainty of context information captured by multiple independent devices, several models to estimate reliability of context information captured in the environment is proposed and developed. Findings -The proposed model has been implemented as a middleware and evaluated using data from real experiments in various scenarios and environment settings. The results of the experiments show the robust performances of the proposed model as the basis for situation reasoning in the environment. Originality/value -A novel model to represent situations and context information captured by multiple devices and to estimate reliability context information used for situation reasoning is proposed. The proposed model is then implemented as a middleware and validated using context data taken captured by multiple independent devices in a MP2P environment.

show abstract