Scenario Based Fault Detection in Context-Aware Ubiquitous Systems using Bayesian Networks

We define the use of Bayesian networks for fault detection in the perception mechanism of context-aware ubiquitous systems. This paper describes the complete working of such a fault detection module. context-aware ubiquitous systems use a large number of sensors and actuators for their interaction with the environment. The data collected from the environment describes the behavior of the system under different scenarios. If in any way this data or the source of the data gets corrupted then the context formed from such data would be erroneous and result in over-all system misbehavior. Therefore such data needs to be filtered and possible sources of error should be detected in the system´s perception mechanism