Super-Resolution Algorithms for Spectral Estimation in Cognitive Radio Systems

摘要

Cognitive radio is new technology that uses spectrum more effectively. Spectrum sensing is one of the most challenging problems in cognitive radio systems. There are some sensing methods for indentifying the unused spectrum for opportunistic transmission. This paper develops one method based on multiple signal classification (MUSIC) algorithm for spectral estimation in cognitive radio system. Unlike traditional MUSIC algorithm that estimates direction of arrival of signal of interest, the modified MUSIC algorithm based on frequency difference of arrival (FDOA) estimates frequencies or carriers of users in cognitive radio network. The developed algorithm brings a more coherent spectrum and real power values than traditional MUSIC as well as fast fourier transform (FFT) approach. In this paper, we also discuss resolution issue and evaluate parameters having significant effects on it.