Passive TDOA Location Algorithm Based On Neural Network

In the rapid development of Passive location technology today, TDOA(Time Difference of Arrival) has gradually become the mainstream of passive positioning for its anti-interference, high accuracy and easy implement. The principle of this positioning method is to receive signals from the signal source which is carried by the target through fixed base stations and calculate the time difference to determine the location of the target. The important research goal of this article is how to locate the target accurately and fast based on non-linear relationship between the time difference and the target coordinates.The traditional solution method requires the target initial position coordinates with high precision, the amount of computation is large and may not be able to get the final solution. As the neural network model does not ask of the characteristics of the data, therefore it is widely used in the field of nonlinear function approximation. This paper presents a neural network model which is used for time difference for passive TDOA location.First, this paper introduced the research background and significance, described the developments and applications of relevant TDOA method and neural network theory. On the basis of a brief introduction of the overall positioning system, described two models for the positioning in detail based on different acquisitions of time difference, and choose the location model based on Internal Synchronization by comparison analysis. Then, briefly introduced the basic theory of neural networks and algorithm principle of BP network and RBF network which are used in this paper. On this basis, established BP neural network positioning model and RBF neural network positioning model for TDOA, provide the simulation results and network errors. Finally, test the performance of two networks with the ideal test data and test data with deviation. The results show that the positioning accuracy of the network is achieved the desired requirements. Because this passive location based on neural network algorithms are high precision and easy to implement, it has a good prospect of application in TDOA.