A New Method for Multisensor Data Fusion Based on Wavelet Transform in a Chemical Plant

This paper presents a new multi-sensor data fusion method based on the combination of wavelettransform (WT) and extended Kalman filter (EKF). Input data are first filtered by a wavelettransform via Daubechies wavelet db4 functions and the filtered data are then fused based onvariance weights in terms of minimum mean square error. The fused data are finally treated byextended Kalman filter for the final state estimation. The recent data are recursively utilized toapply wavelet transform and extract the variance of the updated data, which makes it suitable to beapplied to both static and dynamic systems corrupted by noisy environments. The method hassuitable performance in state estimation in comparison with the other alternative algorithms. Athree-tank benchmark system has been adopted to comparatively demonstrate the performancemerits of the method compared to a known algorithm in terms of efficiently satisfying signal-tonoise(SNR) and minimum square error (MSE) criteria.

Keywords:

Multisensor , Data Fusion , Wavelet Transform , Extended Kalman Filter , Minimum Mean Square Error (MMSE)