Parametric Space Dimensionality Reduction in Multidimensional Signal Interpolation

Abstract

The reduction of the dimensionality of a parametric space is done in the adaptive interpolation of a multidimensional signal. A hybrid adaptive interpolator underlies the dimensionality reduction. The multidimensional hybrid interpolator uses structurally different algorithms to interpolate multidirectional sections of the signal. The approximation of some sections of the signal by other sections underlies the interrelations between signal sections. The adaptive parametric interpolation of intra-sectional readings accounts for intra-sectional interrelations between signal readings. Computational experiments on real multidimensional signals prove the efficiency of the hybrid adaptive interpolator.