Adaptive Kalman Filtering of Colored Noise

摘要

In this paper, an adaptive filtering approach. of colored noise based on the Kalman filter structure using a single layer neural network is proposed, which is unnecessary to know the noise statistics of the plant beforehand and extend the dimensions of the filter. The colored measurement noise is first modeled from a Gaussian white noise through a shaping filter. The Kalman filtering model of colored noise is then built by adopting an equivalent observation equation, which can avoid the dimension extension and complicated computations. An observation correlation-based algorithm is suggested to estimate the variance of the measurement noise by use of a single layer neural network. The adaptive Kalman filtering approach of colored noise is applied to restore the cephalometric image of stomatology. The experimental result demonstrates the feasibility and good performance of the approach.