An adaptive Kalman filter: application to processing the measurement results of MWC photometer

As an analytical instrument for dynamic absorbance detection of liquid samples, metal-waveguide-capillary (MWC) photometer has wide applications in the field of environmental monitor and life science. However, the measurement results of photometer contain different levels of noise inevitably, which is not conducive to realizing high sensitivity. As a classical data processing method, Kalman filter is widely used in data processing and state estimation of sensors. For the measurement results of MWC photometer, a new adaptive Kalman filtering method is proposed in this paper. Based on the sliding window and forgetting factor algorithm, the adaptive mechanism is introduced by adjusting the noise parameters of Kalman filter dynamically, which can deal with the nonlinear and noise changes in the measurement results of MWC effectively, so as to improve the filtering effect. As a result, the performance of the MWC photometer was improved.