From periodic sampling to irregular sampling through PNS (Periodic Nonuniform Sampling)

Abstract

Resampling is an operation costly in calculation time and accuracy. It regularizes irregular sampling, replacing N data by N periodic estimations. This stage can be suppressed, using formulas built with incoming data and completed by sequences of elements which influence decreases when the number of data increases. Obviously, some spectral properties (for processes) and some asymptotic properties (for the sampling sequence) have to be fulfilled. In this paper, we explain that it is possible to elaborate a logical theory, starting from the ordinary periodic sampling, and generalized by the PNS (Periodic Nonuniform Sampling), to treat more general irregular samplings. The "baseband spectrum" hypothesis linked to the "Nyquist bound" (or Shannon bound) is generalized to spectra in a finite number of intervals, suited to the "Landau condition".