Affiliation:
1. Alexandria University, Egypt
Abstract
In this chapter, the authors discuss how compressive sensing can be used in wideband spectrum sensing in cognitive radio systems. Compressive sensing helps decrease the complexity and processing time and allows for higher data rates to be used, since it makes it possible for the signal to be sampled at rates lower than the Nyquist rate and still be reconstructed with high accuracy. Different sparsifying bases for compressive sensing are presented in this chapter and their performance is compared. The design of these matrices is based on different types of wavelet transforms, including the discrete wavelet transform and the stationary wavelet transform; the latter having shown a clear improvement in performance over the former. The authors present different ways of implementing these transforms in a compressive sensing framework. Additionally, different types of reconstruction methods including the genetic algorithm and the auxiliary function method are also presented and their impact on the overall performance is discussed.
Reference33 articles.
1. Cyclostationary feature based multiresolution spectrum sensing approach for DVB-T and wireless microphone signals
2. Comparative Performance Evaluation of Spectrum Sensing Techniques for Cognitive Radio Networks
3. Compressive Sensing [Lecture Notes]
4. Blair-Stahn, N. (2000). Random Permutation Matrices. Academic Press.
5. Candes, E., & Romberg, J. (2005). l1-magic: Recovery of sparse signals via convex programming. Retrieved from www.acm.caltech.edu/l1magic/downloads/l1magic.pdf