Rotating machinery is widely used in all of the industrials fields. In this sense, mechanical components are common elements in these systems. Therefore a periodic revision of the mechanical components of the system is required in order to guarantee the normal running conditions. Any failure or wear must be detected quickly and repaired before it affects to the machine’s integrity. The solution is making a preventive maintenance based on defect diagnosis. Acoustic emission is a new technology for fault diagnosis in mechanical elements that is gaining acceptance as a useful analysis tool in mechanical components despite of the classical vibration analysis. The aim of this project is the study of fault diagnosis methods, by means of acoustic emission signal analysis, applied to electromechanical systems. This study is made with the aim of developing a pseudo automatic fault detection system into an embedded platform, such a μprocessor board.