Xampling: Signal Acquisition and Processing in Union of Subspaces - Computer Science > Information TheoryReportar como inadecuado




Xampling: Signal Acquisition and Processing in Union of Subspaces - Computer Science > Information Theory - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: We introduce Xampling, a unified framework for signal acquisition andprocessing of signals in a union of subspaces. The main functions of thisframework are two. Analog compression that narrows down the input bandwidthprior to sampling with commercial devices. A nonlinear algorithm then detectsthe input subspace prior to conventional signal processing. A representativeunion model of spectrally-sparse signals serves as a test-case to study theseXampling functions. We adopt three metrics for the choice of analogcompression: robustness to model mismatch, required hardware accuracy andsoftware complexities. We conduct a comprehensive comparison between twosub-Nyquist acquisition strategies for spectrally-sparse signals, the randomdemodulator and the modulated wideband converter MWC, in terms of thesemetrics and draw operative conclusions regarding the choice of analogcompression. We then address lowrate signal processing and develop an algorithmfor that purpose that enables convenient signal processing at sub-Nyquist ratesfrom samples obtained by the MWC. We conclude by showing that a variety ofother sampling approaches for different union classes fit nicely into ourframework.



Autor: Moshe Mishali, Yonina C. Eldar, Asaf Elron

Fuente: https://arxiv.org/







Documentos relacionados