Bootstrapping the spectrum in ultra wide-band cognitive radio networks

Abstract

The ultra wide-band cognitive radio networks, in recent years, have attracted the attention of the research due to their challenges concerning the signals detection over multiband. The aim of this paper is to address the ultra wide-band spectrum sensing, i.e. to detect the existence of spectral holes in ultra-wide frequency bands in order to improve the spectrum utilization and minimize the interference to the primary communications. Because of channel fading, the detection by one cognitive radio (CR) receiver is subject to missed detection, therefore it is instrumental for a network of spatially diverse CRs to collaborate during the spectrum sensing. In this paper, the collaborative detection is performed and the decentralized decision is discussed. A new paradigm for the decentralized decision based on bootstrap sampling in the spatio-frequency domain is presented. The performance of the proposed spectrum sensing approach for ultra wideband cognitive radio networks is also investigated.