Algoritrm Resrarch Of Multiscale Data Fusioin State Estimation

Multisensor data fusion is an emerging technology applied to Department of Defense (DoD) areas such as automated target recognition, battlefield surveillance, and guidance and control of autonomous vehicles, and to non-DoD applications such as monitoring of complex machinery, robotics, and medical diagnosis etc.In principle, fusion of multisensor data provides significant advantages over single source data. In addition to the statistical advantage gained by combining same-source data (e.g., obtaining an improved estimate of a physical phenomena via redundant observations), the use of multiple types of sensors may increase the accuracy with which a quantity can be observed and characterized. Significant investments in DoD applications, rapid evolution of microprocessors, advanced sensors, and new techniques have led to new capabilities to combine data from multiple sensors for improved inferences. Recently however, interest in wavelets has grown at an explosive rate. One of the more recent areas of investigation in multiscale analysis has been the emerging theory of multiscale representations of signals and wavelet transforms and the development of multiscale signal processing algorithms.In this paper, by combining the multiscale representations of signals with data fusion techniques, we describe several algorithms for modeling stochastic phenomena at multiple scales and for their efficient estimation or reconstruction given partial and/or noisy measurements which may also be at several scales. In the end of each algorithm, we give numerical examples to show how our models can be used to smooth noisy data as well as examples of fusing multiscale data. The main contributions of this paper are as follows:1. The introduction of the multisensor data fusion on models and algorithms.2. A new multiscale recursive data fusion estimation algorithm is presented. We prove that it is optimal in the linear minimum variance sense.3. Based on the low-pass filter coefficients of Daubechies 4 wavelet, we present another multiscale recursive data fusion estimation algorithm.4. By combining wavelet transform with Kalman filtering techniques, based on the preprocessing of the multisensor multiscale measurements, we present a new method which can effectively remove noises.