Research On Ultra-wideband Localization

With the rapid development of the Internet of Things(Io T)industry,location services have gradually penetrated into people's lives.For example,in smart agriculture,the Io T collects data such as soil moisture and ambient temperature through sensors,then analyzes these data,and combines location information to achieve precise management of agriculture.Thus precise positioning is the key to the application of Io T location services.The commonly used positioning technologies mainly include beidou satellite positioning,bluetooth positioning,and ultra wideband(UWB)positioning.The UWB positioning has the advantages of simple hardware circuit implementation,low system power consumption,high positioning accuracy,and can improve centimeter-level services.It is widely used in fields such as intelligent robots,intelligent agriculture,and autonomous driving.Therefore,the study of the UWB positioning is of great significance.On the basis of research achievements in the UWB localization algorithm,we improve the cuckoo localization algorithm to further enhance the positioning precision.The traditional cuckoo positioning algorithm only uses part of the distance information,that is,the distance information between the anchor node and the unknown node,and the positioning information is incomplete,which result in a limited improvement in positioning accuracy.In order to improve the accuracy of location estimation,the traditional cuckoo location algorithm is improved,and a new fitness function is established by making full use of the distance information between unknown nodes.Considering the influence of the initial position on the improved localization algorithm,the random position is no longer selected as the initial position,but the good initial position is provided by the traditional cuckoo localization method.The simulation results show that the improved algorithm can further reduce the influence of ranging error on positioning performance.Compared with the traditional cuckoo positioning algorithm,the improved algorithm can improve the positioning accuracy,but its positioning performance is also affected by the non-line-of-sight(NLOS)propagation environment.It need analyze the channel statistical characteristics such as kurtosis,root mean square delay spread,skewness to identify the NLOS status.The prerequisite for realizing NLOS state discrimination is to know the type of channel environment.The deep belief network(DBN)is a deep network structure composed of multi-layer restricted boltzmann machines(RBM)and has a good classification effect.In view of this,this paper proposes an UWB channel environment classification algorithm based on the DBN.First,different UWB channel environment data is sampled to construct a sample data set.Then,the sample data set is used to train a DBN.The trained network can extract different UWB channel environment characteristics,and realize the classification of the channel environment.