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