Research On Cooperative Spectrum Sensing Algorithms In UAV Cluster

With the wide application of UAV in emergency communication,enhanced coverage,remote sensing monitoring,military confrontation and many other fields,UAV cluster communication has become the focus of academic and industrial attention.However,the competitive use mode and random interference characteristics of public open spectrum in the current complex environment cannot guarantee the normal communication of UAV cluster,and the increasing shortage of high-quality spectrum resources restricts the rapid development of UAVs.Spectrum sensing is one of the key technologies to alleviate the shortage of spectrum resources.Considering the noise uncertainty and network topology variation of UAV cluster in complex environments,it is of great significance to deeply study the corresponding efficient spectrum sensing algorithm.In this paper,the collaborative spectrum sensing algorithm of UAV cluster is studied and discussed.The existing spectrum sensing algorithm is classified and summarized,and relevant research progress at home and abroad is tracked.Two spectrum sensing methods are proposed for different application scenarios of UAV cluster communication,and the main research contents are as follows:(1)On the basis of summarizing the spectrum sensing algorithm and UAV cluster communication theory,aiming at the characteristics of UAV cluster communication network topology and application environment,UAV network is divided into two scenarios,and the problems in designing UAV cluster cooperative spectrum sensing algorithm are analyzed.(2)According to the characteristics of the unmanned aerial vehicle cluster network topology unchanged scenarios,analyzes the existing unmanned aerial vehicle cluster in collaborative spectrum sensing algorithm and feature extraction of noise robustness and threshold is complex,the deficiency of the poor based on the measured characteristics and KMeans clustering algorithm,this paper proposes a measure based on the characteristics and improved K-Means clustering algorithm of spectrum sensing.The K-Means clustering algorithm is improved,and the interval of density setting is adjusted according to the density distribution of distance features to avoid the error caused by initial random selection of clustering centers.The simulation results show that the proposed algorithm can effectively improve the cooperative spectrum sensing performance in the environment with few UAV sensing nodes and large noise interference.when the signal-to-noise ratio is greater than 10 d B,the average detection probability of this algorithm can reach 0.99.(3)For unmanned aerial vehicle cluster UAV perceive time node mobility and network topology changes,according to the fixed timeslot spectrum bandwidth resources,waste of collaborative perception algorithm and sequential perception can be adjusted dynamically according to the results of the sensory testing the characteristics of the time slot,this paper studies the maximum-minimum distance clustering based sequential collaborative spectrum sensing algorithm.The simulation results show that based on the clustering data processing and two-step perception strategy combined with UAV motion state,the collaborative perception time of UAV cluster is reduced,and the perceived performance of the current ordered collaborative spectrum perception algorithm is optimized.The maximum-minimum distance clustering data processing and two-step perception strategy are adopted.Compared with random clustering method and maximum node degree clustering method,the average detection probability is more than 0.2.