EFFECT OF PERTURBATION ON ESTIMATION OF A UTOREGRESSIVE MODELS

In spectral analysis and pattern recognition, especially in speech pro cessing, AutoRegressive (AR) modeling is widely used as a signal param eter estimation tool. In this paper, AR estimation behaviour is studie d when applied to signals presenting abrupt changes at an unknown inst ant of time. A particular abrupt change is studied : the additive chan ge case. This change occurs when, at the instant of change, a signal a dds to one which was present before - both signals are independent -. The theoretical sliding window AR parameter expressions are given. We show that blind sliding window AR estimation can lead to a detection t ool, allowing additive changes to be detected and distinguished.