An Adaptive Filter for Aeromagnetic Compensation Based on Wavelet Multiresolution Analysis

Abstract

In aeromagnetic compensation, the multicolinearity of the Tolles-Lawson model is one of the main factors limiting the performance of the coefficient-estimating methods. Aside from the intrinsic characteristic of the Tolles-Lawson model, data filtering as a preprocessing step in estimating the coefficients has a significant impact on the multicolinearity. To solve this problem, an adaptive filter based on wavelet multiresolution analysis is proposed. By modifying the wavelet decomposition results of the Mallat algorithms and reconstructing the signal, the adaptive filter can be used to establish a system of linear equations not only by capturing the frequency characteristics of the aircraft maneuvers as is usual but also by having a weakened multicolinearity. Simulation results illustrate the effectiveness of the proposed adaptive filter.