Spectral Analysis of Nonuniformly Sampled Data Using Sparse Signal Representation
Zhen-Qing He, Qinghua Liu · International Conference on Electric Information and Control Engineering · 2012
For a given finite set of nonuniformly or irregularly sampled data, we present a new approach based on sparse signal representation (SSR) framework for their spectral analysis, which is capable of significantly achieving a higher resolution over the various existing methods: Schuster periodogram and iterative adaptive approach (IAA). In this paper, the observed data of nonuniformly sampling can be cast as a single measurement vector (SMV) model via over complete sparse signal representation, and then the spectrum is estimated by recovering a sparse vector. The simulation results confirm the efficacy of this proposed approach.