Extraction Algorithm of Unit Vibration Characteristic Signals Using Independent Component Analysis Method

To solve the problems of unit noise and vibration causing inconvenience to residents, this paper proposes a unit noise and vibration device and designs a unit noise and vibration control system. Adaptive active noise and vibration control is adopted, and DSP is used as the core of the entire system, which includes the acoustic sensor part, the peripheral signal conditioning part and the signal processing control part, so that it can achieve real-time control of unit noise and vibration. In addition, the improved independent component analysis method is used to achieve the noise reduction effect of unit noise and vibration. Finally, a unit noise and vibration detection system are designed, MEMS microphone detection is applied to achieve real-time detection of unit noise and vibration. Experiments show that the noise reduction effect of the system studied in this paper is relatively good. The vibration effect is better when the distance from the unit is 1m, and the noise value is reduced from 67[Formula: see text]dB to 48[Formula: see text]dB. Meanwhile, when the vibration distance from the unit is 22[Formula: see text]m, the noise drops to 37[Formula: see text]dB, and the noise value detected by the system in this paper is not much different from the actual noise value.