Unknown Source Localization Of Urban Grid System Based On Artificial Neural Network

Wireless localization is a kind of technology which used for calculating the unknown source location through direct or indirect determination of radio signal at a particular location and time,in the process of transmission between fixed monitoring the change of phase,amplitude and frequency.It is important for the field of military,navigation,environmental control,road navigation,tracking property.Optimize positioning technology,making the result more accurate means a lot as far as national security,science and technology development and the improvement of people’s life quality.Artificial Neural Network(Artificial Neural Network,ANN)has been proposed in the early 20 th century,human understanding of biological neural network on the basis of artificial structure can implement some functions of the Network.Nowadays,artificial neural network has been used in computer science,automatic control science,psychology,fault diagnosis and detection and economics and other fields.It has been fully combined with the actual work of Tianjin radio stations wireless monitoring content,in order to raise the accuracy in finding unknown source with urban grid system,this paper proposed a method of using artificial neural network(ANN)to locate.Using BP neural network and RBF neural network to locate in the study area,we compared the two methods.What’s more,in order to get more accurate positioning effect,we used the genetic algorithm(GA)to optimize RBF neural network.Based on city grid system for unknown source localization,in the use of field intensity distribution information to reduce the search range for accurate positioning,after that using neural network to get the location information,this method can compensates the problem when using the system built-in TDOA location station big workload and the problem of low accuracy.The main innovation points is to radio stations in Tianjin city grid system as the foundation,using the artificial neural network for unknown source localization,and optimized the RBF neural network using genetic algorithm.Through the simulation of an outdoor scene,we selected a few stations in the scene to monitor,and found the coordinates of the unknown source and transmission power more accurate.And verifying the method is feasible,it can effectively solve practical problems.