Valery Kanevsky

Bayesian fault diagnosis in large-scale measurement systems

Because of both increasing complexity and increasing geographic distribution, faults in measurement systems are becoming more troublesome to diagnose in all life-cycle phases: manufacturing, deployment, and operation. This paper considers which features are necessary in an automatic diagnosis system for large-scale measurement systems. The paper describes MonteJade, a diagnosis system designed with these essential features in mind. Examples are given of MonteJade diagnosing faults and determining the next debugging actions to perform, in a measurement system constructed from more than one hundred components.

Learning from Randomly-Distributed Inaccurate Measurements

Traditional measurement systems are designed with tight control over the time and place of measurement of the device or environment under test. This is true whether the measurement system uses a centralized or a distributed architecture. Currently there is considerable interest in using mobile consumer devices as measurement platforms for testing large dispersed systems. There is also growing activity in developing concepts of ubiquitous measurement, such as “smart dust.” Under these conditions the times and places of measurement are random, which raises the question of the validity and interpretation of the acquired data. This paper presents a mathematical analysis that shows it is possible under certain conditions to establish dependence between error bounds and confidence probability on models built using data acquired in this manner.