Research On The Technology Of Energy Detection Based Adaptive Spectrum Sensing Algorithm

With the rapid development of wireless communication technology and business, how to effectively use the limited spectrum resources to meet the more and more radio service demands of people has become a widely concerned matter. Spectrum sensing technology is considered as one of the key technologies to solve the problem of the lower spectrum utilization rate, is also one of the most popular technology that attracts the most attention in the world. Spectrum sensing process, monitoring the electromagnetic spectrum around to find out spectrum holes and assigning it to non-licensed users, can make the non-licensed user to use the spectrum opportunisticlly, which greatly improves the spectrum efficiency. Spectrum sensing technology has been widely applied in many fields, such as military communication, civil communication, public services, medical enterprise and so on. It has gained great attentions form academia and industry due to its broad application prospect and high commercial value.This manuscript firstly introduces the background, overview, the main challenges and the research status of cognitive radio and spectrum sensing technology. Then, according to the distortion of the received signal energy caused by the channel mutations, we propose an efficient and robust adaptive spectrum sensing algorithm based on energy detection technology. This algorithm adopts the method of adaptive parameter instead of the traditional soft collaboration performing the signal detection. We design different cost functions for single node signal detection and weighted cooperative signal detection respectively. To obtain a new test statistics, linear minimum mean square error algorithm is utilized to solve the mean square error minimization problems of these cost functions. In addition, the proposed algorithm greatly improves the deflection coefficient, thus improving the detection performance. Moreover,the performance of the algorithm were evaluated through theoretical derivation and simulations.Due to impossible to get noise energy in practical, an energy-eigenvalue joint adaptive detection algorithm and a general model of continuous multistage detector are proposed in chapter IV. This algorithm belongs to completely blind detection algorithm, and it has a compromise between the detection accuracy and complexity. This proposed algorithm firstly estimate the noise variance by the second stage detector, and then feedback it to the first stage detector to make a completely blind adaptive detector. The impact of signal bandwidth and observation bandwidth for this adaptive detector has been studied along with the optimal detection bandwidth. Finally, the complexity of the proposed joint energy eigenvalues detector is analyzed in this paper, which is also compared with the traditional energy detector and eigenvalue detector.Consequently, abundant simulations verify that the proposed adaptive energy detection algorithm can highly improve the detection performance. Without the prior information of primary user, extensive simulations are performed to demonstrate the efficiency of the energy-eigenvalue joint adaptive detection algorithm, which can meet the spectrum sensing requirements of the mobile communication system.