Wideband spectral estimation from compressed measurements exploiting spectral a priori information in cognitive radio systems

In Cognitive Radio scenarios channelization information from primary network may be available to the spectral mon-itor. Under this assumption we propose a spectral estima-tion algorithm from compressed measurements of a multi-channel wideband signal. The analysis of the Cramer-Rao Lower Bound (CRLB) for this estimation problem shows the importance of detecting the underlaying sparsity pattern of the signal. To this end we describe a Bayesian based iterative al-gorithm that discovers the set of active signals conforming the band and simultaneously reconstructs the spectrum. This iter-ative spectral estimator is shown to perform close to a Genie-Aided CRLB that includes full knowledge about the sparsity pattern of the channels. Index Terms — Spectral estimation, compressive sam-pling, cognitive radi