Selective Cooperative Spectrum Sensing

As one of the most promising dynamic spectrum access technologies, cognitive radio(CR) technology can solve the conflict between the shortages of spectrum resource andunderused licensed spectrum bands. The core idea of CR technology is sharing licensedspectrum without interfering with primary users. Thus one of the most important functionsof CR is to detect idle spectrum bands in its radio environment, i.e, spectrum sensingtechnology. The essence of spectrum sensing is differentiating the signal of primary usersfrom noise. In general there is no prior knowledge of the signal and locations of primaryusers. On the other hand, primary users will not share their licensed spectrum withcognitive users with inferior sensing performance. For above factors, conventional signaldetection methods can not be applied in CR systems directly. New theoretical methods andkey technologies are needed. The major contributions of the dissertation are as follows.1. A fast and efficient spectrum sensing method based on cyclic autocorrelation functionis proposed.Cyclic autocorrelation (CA) characteristics of the digital modulated signal are analyzedin this paper. A novel spectrum sensing method based on CA is proposed. Spectrumsensing based on nonzero cyclic frequency and the zero cyclic frequency is analyzed.Detection probability and false alarm probability of spectrum sensing are derived relyingon theoretical analysis. When cyclic frequency is α=0, R x (0,τ≠0)has maximumvalue and so it has the best detecting performance. Simulation results show that theproposed method has anti-noise advantage based on cyclostationary characteristics and lowimplementation complexity. The computational cost is far less than the cyclic spectrumestimation and is close to the energy method. The theoretical analysis and simulationresults have proved that this is an effective fast spectrum sensing method.2. A selective cooperative sensing strategy and a user selection method are proposed.To reduce sensing overhead and total energy consumption, it is recommended that the users with good performance should be selected to increase the sensing reliability.Considering the difference of cognitive nodes in sensing performance, a cooperative nodesselection scheme based on the individual characteristic is proposed. A selective cooperativesensing strategy and a user selection method are also proposed so as to increase the sensingreliability and reduce sensing overhead.3. The spectrum sensing based on the multi-antenna selection fusion is realized.Multi-antenna spectrum sensing based on the likelihood ratio needs to have the priorknowledge of the the channel gain, the noise variance and the primary user’s signalingscheme. Multi-antenna spectrum sensing based on the generalized likelihood ratiotest (GLRT) overcomes these limitations, but the calculation of maximum likelihoodestimation is complex. The author puts forward EGC、MRC and SC three fusion sensingmethods based on CA. The performance analyses for the proposed methods are presentedand the expressions of detection probability are also derived. Simulation results show thatthe proposed EGC fusion sensing exhibits better performance than that of the GLRT-basedsensing.Using all antennas in the network does not always achieve the best sensing performance.In this paper, a antenna performance parameterAUi and a selection parameter A Uiareproposed. A Uiis used to select the well-performed antenna diversity to fusion sense.The simulation results indicate that the proposed antenna selection fusion algorithm is ableto optimize network performance.