Research On Efficient Spectrum Sensing Technology In Cognitive Radio

In recent years,with the rapid development of the communications industry,more and more networks using new wireless technologies have been deployed on a large scale.As a result,a variety of networks coexist,and there are fewer and fewer spectrum resources available for distribution.In addition,the fixed allocation of spectrum resources makes the problem of shortages even more serious.The spectrum sensing technology in cognitive radio can monitor spectrum holes in real time,and provide a data basis for secondary users to use dynamic spectrum access technology to access these holes for communication.Therefore,it has been widely studied in recent years.This thesis mainly studies the spectrum sensing problem in various application scenarios.The main work is as follows:In the narrow-band spectrum sensing scenario,this thesis studies energy sensing technology,Shannon entropy and exponential entropy-based spectrum sensing technology,and beamforming-based fast spectrum sensing technology under the background of millimeter wave.And simulates and analyzes their characteristics,performance,and scope of application.Based on the scenario of a two-layer heterogeneous network,this thesis studies the two-layer cooperative spectrum sensing technology based on dual-threshold,the twolayer cooperative spectrum sensing technology based on noise uncertainty,and the cooperative dual-spectrum sensing technology based on dynamic dual-threshold.Compared with traditional dual-threshold cooperative spectrum sensing technology,it improves the robustness to noise power uncertainty.In addition,for the case where the channel conditions of each secondary user are different,a cooperative spectrum sensing technology based on improved K rank and a cooperative spectrum sensing technology based on selected users are proposed.Compared with traditional collaborative sensing technology,these two algorithms significantly improve the performance of the system.Based on wideband spectrum sensing technology,this paper studies the basic compressed sensing technology.Based on this,a sequential broadband sensing technology based on packet-compressed data is proposed,which solves the problem of excessive computing resource consumption when the existing algorithms continuously observe the broadband spectrum.On top of this,we continue to improve,and propose a broadband fast-change sensing technology to further reduce the detection delay of broadband signals and improve system performance.In addition,this paper also studies the detection of spread spectrum signals.Based on the traditional cyclic auto-correlation-based spread spectrum signal detection,an improvement is made,this thesis uses the second-order moment of the signal cyclic autocorrelation as the detection target quantity,improves the detection performance of the system;and the use of Wiener Sinchen's theorem reduces the computational complexity of the algorithm.In addition,this thesis also studies the direct spread signal detection technology based on the peak-average power ratio,which greatly improves the robustness of the algorithm to the uncertainty of noise power.