Research On Dynamic Wide-band Spectrum Sensing Algorithm Based On Compressive Sensing

With the development of wireless communication technology,the demand for spectrum resources is increasing more and more.Radio frequency spectrum become a valuable but tightly regulated resource due to its unique and important role in wireless communications.However,recent study reveals that the spectrum utilization is very low due to the static spectrum management.Cognitive radio(CR)was first proposed in 1999,and it has become a promising solution to make up for the spectrum scarcity.In wide-band spectrum sensing,the signal have sparsity in frequency domain because that the utilization of wide-band spectrum is very low.So we can introduced Compressive sensing(CS)to wide-band spectrum sensing.The most of present wide-band spectrum sensing algorithms need to reconstruct the whole spectrum or Power Spectral Density(PSD)of the original signal,and use it to detect the spectrum occupancy status.Furthermore,wide-band spectrum sensing encounters many challenges from the wireless network environment,such as wireless fading environment and hidden terminal problem,those problems can harm the detection performance of wide-band spectrum sensing.To address these and several problems associated with spectrum sensing,we propose a dynamic compressive wide-band spectrum sensing algorithm and a collaborative dynamic compressive wide-band spectrum sensing algorithm.The major researches contents of the thesis are as follows:Putting forward a dynamic compressive wide-band spectrum sensing algorithm based on channel energy reconstruction in cognitive radio networks.Taking a more practical network scenario in which each channel changes occupancy status with a small probability into account.By using the sparsity of the channels' energy,a wide-band random filters bank is employed to achieve channel energy measurements.The proposed algorithm can detects the changes of channel's occupancy status by changes of measurements before and after,then updates the channel's energy vector dynamically.The proposed algorithm can recover the energy of each channel that is changing occupancy status in consecutive time slot directly rather than recovering the whole spectrum.The simulation results show a significant improvement in probability of detection and probability of false alarm,in comparison to traditional algorithm.In addition,simulation results show the fast speed and robustness to noise of theproposed algorithm.Putting forward a collaborative dynamic compressive wide-band spectrum sensing algorithm based on channel energy reconstruction in cognitive radio networks.To address the hidden terminal problem that brought by wireless fading environment,this thesis has proposed a collaborative version of the dynamic compressive wide-band spectrum sensing algorithm.By using the characteristic that the channel energy vector of different SUs have same sparse support,the data fusion center will do the fusion processing of multiple measurements form SUs with collaborative method,and it can reserve the effective information at the greatest extent by executing the elimination processing.The simulation results show that the collaborative sensing can improve the performance of spectrum sensing,and the more number of SUs that participate in the collaboration,the better the performance of spectrum sensing.Moreover,the proposed algorithm can overcome the effects of wireless fading environment by collaborative sensing,and avoid the sensing deficiencies due to the single SU sensing.