ABSTRACT
Cognitive radio (CR) has been acknowledged as a probable solution to combat spectrum shortage problem because of diverse wireless applications. Cyclostationary (CS) feature detection is a spectrum hole (i.e. vacant licensed frequency band) detection technique with higher degree of accuracy than the energy detection system. This method takes advantage of time varying autocorrelation function. Higher order CS detector usually exploits more than second-order probabilistic functions of CS signals. In this paper, we propose to detect spectrum holes in multi-antenna-assisted CR using higher order cyclostationarity. The entire detection has been performed under consideration of Gaussian as well as non-Gaussian noise (Gamma distribution). In this context, mutual coupling effects in the receiver section have been considered. Finally, the expressions for false alarm as well as detection probabilities are derived. Numerical results are given in support of the proposed detection technique.
Disclosure statement
No potential conflict of interest was reported by the authors.