Title :
A compressive sensing recovery algorithm based on sparse Bayesian learning for block sparse signal
Author :
Wang Wei ; Jia Min ; Guo Qing
Author_Institution :
Commun. Res. Center, Harbin Inst. of Technol., Harbin, China
Abstract :
Compressive sensing offers a new wideband spectrum sensing scheme in cognitive radio. In this paper, a sparse signal recovery algorithm based on sparse Bayesian learning (SBL) framework is proposed. By exploiting intrablock correlation in a block sparse model and using Expectation-Maximization (EM) method, this algorithm achieves superior performance. The results of experiments show that this algorithm is robust to noise and has better performance than other algorithms in signal recovery. Then we apply it to wideband spectrum sensing, we find that proposed algorithm not only guarantees accurate signal estimation, but also obtains higher correct detection probability.
Keywords :
cognitive radio; compressed sensing; expectation-maximisation algorithm; learning (artificial intelligence); radio spectrum management; signal detection; EM method; SBL framework; block sparse signal; cognitive radio; compressive sensing recovery algorithm; detection probability; expectation maximization method; intrablock correlation; signal estimation; signal recovery; sparse Bayesian learning; sparse signal recovery algorithm; wideband spectrum sensing; wideband spectrum sensing scheme; Bayes methods; Compressed sensing; Correlation; Sensors; Signal processing algorithms; Signal to noise ratio; Wireless communication; Compressive sensing; intra-block correlation; signal recovery algorithm; sparse Bayesian learning (SBL); wideband spectrum sensing;
Conference_Titel :
Wireless Personal Multimedia Communications (WPMC), 2014 International Symposium on
Conference_Location :
Sydney, NSW
DOI :
10.1109/WPMC.2014.7014878