Detection of average jump by bayes test with adaptive threshold

Abstract EN

The subject of this paper is to show how to implement a binary hypothesis testing problem ht test would allow the detection of low amplitude faults that affect a signal, as fast as possible ant, with a minimum risk of errors.

It is a Bayes test of adaptive threshold based on the estimation oj c priori probabilities through the Gaussian Kernels method.

A comparative study is provided t < show that such a test improves performance (the increase of detection probability and thi reduction of false alarm probability) compared with Bayes's fixed threshold test.