ARIMA Modeling of Noise of Piezoelectric Accelerometer

Piezoelectric accelerometer is well-known for ultra-high sensitivity that is limited by the noise floor. Due to its importance, the modeling of the noise of piezoelectric accelerometers using Autoregressive Integrated Moving Average (ARIMA) is presented herein. This method is applied to noise data from five commercial and prototype devices with different noise patterns and bandwidths. The best-fit model for each accelerometer is selected after an iterative process. The resulting models are found to be in good match with the original noise patterns. The use of ARIMA to produce an analytical equation of the noise characteristic is helpful for users of commercial piezoelectric accelerometers, as it does not require knowledge of the internal structure of the device that is needed by physics-based noise models.