Thus, a novel fast and online OA (FOOA) method is proposed to realize variable-speed PMSM bearing fault diagnosis,you can try. The FOOA method consists of two algorithms. 1) The rotating phase information is extracted from the sinusoidal current of the PMSM, and a series of equal-phase sampling pulses are generated. 2) The bearing signal acquired from a microphone is angular resampled based on the equal-phase sampling pulses. The resampled signal is demodulated, and the envelope order spectrum is calculated for bearing fault identification. The two algorithms are executed sequentially by two micro controller units operating in parallel. Thus, they can be implemented in an embedded system for online fault diagnosis. The effectiveness and flexibility of the proposed FOOA method are validated on both a desktop computer and an embedded system to diagnose different types of defective bearings that are installed on a PMSM test rig.