Detecting Faults in Context-Aware Adaptation

Clc Number:

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments

    Internetware applications are context-aware. They adapt their behavior based on environmental changes. However, faulty adaptation may arise when these applications face unanticipated situations. Such adaptation faults can be difficult to detect at design time. One promising approach is to statically analyze model-based context-aware applications exhaustively for all potential faults. However, it suffers from expressiveness and precision problems. To address these limitations, we propose in this paper a dynamic adaptation model (AM) approach. AM offers increased expressive power to model complex adaptation rules, and guarantees soundness in its fault detection. In addition, AM deploys an incremental rule evaluation (IRE) technique to cater for context-aware applications, such that it can effciently handle environmental changes in its fault detection. We evaluated AM using both simulated and real-world experiments with two context-aware applications. The experimental results confirmed that AM can detect real faults missed by existing work, and avoid numerous false warnings that were misreported otherwise.

    Cited by
Get Citation

Chang Xu, S. C. Cheung, Xiaoxing Ma, Chun Cao, Jian Lv. Detecting Faults in Context-Aware Adaptation. International Journal of Software and Informatics, 2013,7(1):85~111

Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
  • Received:
  • Revised:
  • Adopted:
  • Online: April 03,2013
  • Published: