On Sensor Sampling and Quality of Information: A Starting Point

  • Authors:
  • Chatschik Bisdikian

  • Affiliations:
  • IBM T.J. Watson Research Center, USA

  • Venue:
  • PERCOMW '07 Proceedings of the Fifth IEEE International Conference on Pervasive Computing and Communications Workshops
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Many sensor-based applications (re)act following the detection of certain events of interest. Hence, the effectiveness of these applications depends on the quality of the information (QoI) provided by their sensor-based event detectors. In this paper, we derive relationships between the QoI attributes of timeliness and confidence and the operational characteristics of sensor systems and the events they detect. By building upon the Neyman-Pearson hypothesis testing procedure, we study the dependence of these characteristics and attributes on each other and establish their theoretical performance boundaries.