Information theoretic criterion-based spectrum sensing for cognitive radio

Abstract

Energy detection is a promising candidate for spectrum sensing of cognitive radio systems due to its merits of simplicity and easy implementation. However, energy detection needs to subjectively pre-configure a threshold and requires the knowledge of noise power. Furthermore, energy detection is vulnerable to noise uncertainty, which inevitably occurs in practical implementations. To solve the above problems, two spectrum sensing methods requiring no knowledge of noise power and no subjective threshold settings are proposed based on the information theoretic criterion. Additionally, the most important advantage of these methods is that their performance is totally unaffected by noise uncertainty. Finally, simulation results are provided to demonstrate the benefits of these methods.