Automatic fault monitoring techniques in machinery are a means to safely re
lax and simplify expensive preventive maintenance and inspection procedures
. Acoustic emissions that are ultrasonic waves emanating from the formation
or growth of a crack in a material is a possible avenue for detecting faul
ts due to fatigue or other sources of deterioration. In this paper we prese
nt some of the signal processing techniques that we have proposed and evalu
ated for providing reliable fault monitoring and diagnostics based on acous
tic emissions.