On the representation of band limited functions using finitely many bits

Authors:
H. N. Mhaskar
Affiliations:
Department of Mathematics, California State University, Los Angeles, California
Venue:
Journal of Complexity
Year:
2002

Citing 3
Cited 5

Selected topics in approximation and computation

Selected topics in approximation and computation
Spherical Marcinkiewicz-Zygmund inequalities and positive quadrature

Mathematics of Computation
On simple oversampled A/D conversion in L2(R)

IEEE Transactions on Information Theory

A tribute to Géza Freud

Journal of Approximation Theory
When is approximation by Gaussian networks necessarily a linear process?

Neural Networks
Polynomial operators and local smoothness classes on the unit interval

Journal of Approximation Theory
Bandwidth of mechanical systems and design of emulators with RBF

Neurocomputing
Functions bandlimited in frequency are free of the curse of dimensionality

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we consider the question of representing an entire function of finite order and type in terms of finitely many bits, and reconstructing the function from these. Instead of making any further assumptions about the function, we measure the error in reconstruction in a suitably weighted Lp norm. The optimal number of bits in order to obtain a given accuracy is given by the Kolmogorov entropy. We determine this entropy in the case of certain compact subsets of these weighted Lp spaces and obtain constructive algorithms to determine the asymptotically optimal bit representation from finitely many samples of the function. Our theory includes both equidistant and non-uniform sampling. The reconstructions are polynomials, having several other optimality properties.