A Cyclostationary Based Spectrum Sensing for UWB and Wi-MAX Coexistance with Noise Uncertainty

This paper focuses on a cognitive radio approach based on a noncooperative spectrum-sensing technique that is suitable for applications in detection and avoidance (DAA) schemes to allow coexistence between ultrawideband multiband (secondary user) and Worldwide Interoperability for Microwave Access (WiMAX; primary user) systems and, hence, improve spectral efficiency. In particular, the cyclostationary property of WiMAX signals due to the cyclic prefix is exploited here. The spectrumsensing method devised in this paper is a statistical test that relies on cyclostationarity, and it belongs to the class of the single-cycle detector. The performance of the proposed detector, in terms of primary-user detection probability, is derived by computer simulations. The results obtained highlight that the proposed approach enables efficient downlink detection under conditions of low signal-to-noise ratio (SNR) and noise uncertainty. The implementation complexity of the proposed method is also discussed here to highlight advantages with respect to previously proposed alternatives. Finally, an analytical approach is presented to derive the false-alarm probability to allow the implementation of a constant false-alarm rate detector. In addition, the use of an efficient interference-mitigation scheme is also considered to improve the bit-error-rate performance.