Shannon-like Interpolation with Spectral Priors and Weighted Hilbert Spaces: Beyond the Nyquist Rate

TL;DR

This paper introduces novel Shannon-like interpolation methods that incorporate spectral priors through weighted Hilbert spaces, enabling improved signal reconstruction especially from sub-Nyquist samples, with theoretical insights and practical demonstrations.

Contribution

It proposes new interpolators based on weighted Hilbert spaces that account for nonuniform spectral densities, extending classical Shannon interpolation beyond the Nyquist rate.

Findings

Effective interpolation of sub-Nyquist data.

Improved reconstruction accuracy with spectral priors.

Practical implementation strategies demonstrated.

Abstract

In this work, we draw connections between the classical Shannon interpolation of bandlimited deterministic signals and the literature on estimating continuous-time random processes from their samples (known in various communities under different names, such as Wiener-Kolmogorov filtering, Gaussian process regression, and kriging). This leads to the realization that Shannon interpolation can be interpreted as implicitly expecting that the unknown signal has uniform spectral density within its bandwidth. However, in many practical applications, we often expect more energy at some frequencies than at others. This leads us to propose novel Shannon-like interpolators that are optimal with respect to appropriately-constructed weighted Hilbert spaces, where weighting enables us to accommodate prior information about nonuniform spectral density. Although our new interpolants are derived for data obtained with any sampling rate, we observe that they are particularly useful for interpolating sub-Nyquist data. In addition to theory, we also discuss aspects of practical implementation and show illustrative examples to demonstrate the potential benefits of the approach.