Research On Spectrum Detection Methods In Cognitive Radio Systems

This paper concentrates on the spectrum detection methods in cognitive radio systems. Via spectrum detection, vacant spectrum is reutilized for data transmission in cognitive communication systems,which brings a remarkable enhancement of spectrum utilization. Thus, as one of the key technologies, spectrum detection has always been the focus of academic research. The objective of this paper is to improve the detection precision of licensed spectrum.First, though matched filter detection is claimed to achieve the best performance in terms of detection precision in state-of-the-art works, it is usually unfeasible in practical application due to its stringent requirements on priori information of user's signal. In theory, power spectrum profile of additive white Gaussian noise is linear without fluctuating characteristics while power spectrum profile of received signal from primary user is dynamically fluctuating. By employing this feature, a low-complexity detection method based on the feature of power spectrum fluctuation is proposed in this paper. The proposed method could acquire the high detection probability with a much looser requirement on SNR and sampling points. We first prove the equivalence between the power spectrum obtained by classical power spectrum estimation method and periodogram method, which provides theoretical support for the proposed detection method. Furthermore, the relationship of theoretical threshold with sampling points, noise power and false alarm probability is obtained through the theoretical analysis and derivation. Simulation results show that the proposed method can obtain higher detection probability under low SNR with less sampling numbers.Second, according to the periodic features of radar signal, we propose a periodicity detection mechanism, referred to as "detection and idle",which is capable of determining the presence of radar signals as well as type of radar. The range of detection time and detection period is offered through a theoretical analysis to guarantee sufficient utilization of radar band spectrum with high detection precision and low energy consumption.