Faults Diagnosis through Genetic Matching Pursuit

Signals that carry information regarding the existing defects or possible failures of a system are sometimes difficult to analyze because of various corrupting noises. Such signals are usually acquired in difficult conditions, far from the place where defects are located and/or within a noisy environment. Detecting and diagnosing the defects require then quite sophisticated methods that are able to make the distinction between noises encoding the defects and another parasite signals, all mixed together in an unknown way. Such a method is introduced in this paper. The method combines time-frequency-scale analysis of signals with a genetic algorithm.