Diagnosing and correcting system anomalies with a robust classifier

If a robust statistical model has been developed to classify the “health” of a system, a well-known Taylor series approximation technique forms the basis of a diagnostic/recovery procedure that can be initiated when the system´s health degrades or fails altogether. This procedure determines a ranked set of probable causes for the degraded health state, which can be used as a prioritized checklist for isolating system anomalies and quantifying corrective action. The diagnostic/recovery procedure is applicable to any classifier known to be robust; it can be applied to both neural network and traditional parametric pattern classifiers generated by a supervised learning procedure in which an empirical risk/benefit measure is optimized. We describe the procedure mathematically and demonstrate its ability to detect and diagnose the cause(s) of faults in NASA´s Deep Space Communications Complex at Goldstone, California