Research on Error Compensation Technology of MEMS Accelerometer

The random error of Micro Electro Mechanical Systems (MEMS) accelerometer is one of the important factors affecting its measurement accuracy. This study begins to adjust for the random error of a MEMS accelerometer in order to completely increase its use value. Firstly, Allan variance method is introduced to analyze and quantify each noise term causing random error. The linear relationship of sample output data is determined through a series of steps such as data preprocessing, randomness test, identification model, judgment order and calculation parameters. Accordingly, the Kalman filter is made to lessen the random drift error noise component that affects the measurement accuracy of MEMS accelerometers.