Research On Wide-band Spectrum Sensing For Cogntive Radios

With the rapid development of wireless communication technology, the number of wireless applications and services has growth rapidly, spectrum resource has become a scarce resource. However, the method that spectrum is allocated to specific user regularly causes enormous waste of spectrum resource. Cognitive Radio (CR) technology, has effectively solved the problem. Spectrum sensing technology is the foundation of CR systems, Secondary users (SU) uses spectrum sensing to find the "spectrum holes", under the condition of not affecting the Primary User (PU) use in the authorized spectrum.At present most of the spectrum sensing methods are using the Nyquist sampling theorem, sampling with twice the highest signal frequency, when using broadband spectrum perception, huge amount of data is difficult to handle and hardware is difficult to achieve. Compressed Sensing (CS) theory is a method that reconstruct the original signal using lesser sampling value, the sampling values is lesser than conventional method, which could samples at a relatively low frequency and can reconstruct the signal with high probability. Broadband spectrum exists in a large number of underused spectrum resources, which is the prerequisite, has obvious spectrum sparse, so CS can be used in the wideband spectrum sensing as well.This article researches on broadband spectrum sensing technology. Fist of all, the basic method and theory of spectrum sensing and compressed sensing is introduced, compressed sampling and reconstruct using modulated wideband converter. Aiming at the sparse characteristic is difficult to determine in wireless environment, the Sparse Adaptive Matching Pursuit algorithm and its improved algorithm are studied, accurately recovering original signal in the case of unknown signal sparse degree. Then it improves conventional cooperation spectrum sensing model, sets different weights according to reliability of SUs, so the threshold of energy detection changes dynamically. This article introduces compressed sensing technology in cooperation spectrum sensing improved model, establishes broadband spectrum sensing system model. Various SUs use MWC to compressed sampling for broadband signal,to compressed sampling and reconstruct the signal, to determine each subband signals take up case according to energy detection method, and the performances are simulated. Simulation results show that this algorithm can use lower rate to sample signals and accurately reconstruct the original signal, detect of spectrum occupancy situation and realize the broadband spectrum perception.