Reconstruction of irregularly sampled discrete-time bandlimited signals with unknown sampling locations

The purpose of this paper is to develop methods that can reconstruct a bandlimited discrete-time signal from an irreg- ular set of samples at unknown locations. We define a solution to the problem using first a geometric and then an algebraic point of view. We find the locations of the irregular set of samples by treating the problem as a combinatorial optimization problem. We employ an exhaustive method and two descent methods: the random search and cyclic coordinate methods. The numerical simulations were made on three types of irregular sets of locations: random sets; sets with jitter around a uniform set; and periodic nonuniform sets. Furthermore, for the periodic nonuniform set of locations, we de- velop a fast scheme that reduces the computational complexity of the problem by exploiting the periodic nonuniform structure of the sample locations in the DFT.

Published in:
IEEE Transactions on Signal Processing, 48, 12, 3462-3471
Other identifiers:

 Record created 2005-04-18, last modified 2018-03-17

Download fulltext

Rate this document:

Rate this document:
(Not yet reviewed)