Researches On Tracking Algorithm Of Mobile Station In Wireless Cellular Networks

Wireless location technology is the guarantee of traffic safety and human activities in the military. In modern society, it is playing the more and more major role. At the same time, users are eager to enjoy location services, such as vehicle, logistic management, location query and so on. In the cellular location system, has been located for mobile terminals is usually an ordinary terminal (mobile phone, etc.), which objectively requires the multiple base station equipment measuring radio signal parameters from the mobile terminal by attachments facilities, such as the propagation time, time difference, the signal field intensity, phase, or angle of arrive, etc. And then, through the appropriate location algorithm calculate the approximate location of mobile terminals. Obviously, due to the mobile communication channel noise, multipath disseminating and other adverse factors, cellular wireless location system is difficult to achieve a higher positioning accuracy, location coverage is also be limited by coverage restriction of cellular mobile communication systemWith the evolution of the 3G mobile communication systems, some new technologies of 3G provide new methods and means for the mobile station's location. How to provide the service of mobile users, we need to research and analyze a variety of location methods and programs, simulate them in the Gaussian channel and the actual channel, and evaluate the performance of the algorithm, then propose a new algorithm which more suitable for wireless location in cellular network and has a good performance. In addition, in order to achieve continuous and real-time location, mobile station location needs to be turned to dynamic tracking location. So, many governments and companies devote much manpower and material resources to research and trial-manufacture of the wireless cellular location system.This paper mainly deals with wireless cellular localization algorithms, which are based on time measure values and angle measure values. First of all, several major methods for wireless cellular localization are introduced in this paper. The detailed localization formulae are presented for TDOA system. At same time, several channel models suitable for mobile localization and several kinds of localization errors express that have established the foundation for the localization simulation and the improvement of algorithm are introduced.Then, on the basis of analyzing present localization technology and algorithm, this paper puts its emphases on TDOA and AOA.'A TDOA location algorithm of Taylor series expansion based on Rwgh'is proposed, The Taylor series expansion algorithm is improved in that. Use located results of the Residuals Weighted algorithm as the initial values of Taylor method, and utilize the minimum Residuals instead of the mean square error in the TDOA which is not easy to obtain in practice. The simulation results show that the effectiveness of NLOS error mitigation in location estimate, and has good stable characteristics.We else propose those algorithms:wireless localization algorithm based on the BP neural network, TDOA localization algorithm based on the BP neural network, AOA location algorithm based on the RBF neural network, TDOA/AOA localization algorithm based on the RBF neural network. The first algorithm utilize neural network to locate the mobile position directly, other kinds of algorithms described above all utilize the neural network to correct to NLOS error first, then utilizes the corresponding localization algorithm to carry on the localization. The simulation results show its location accuracy is significantly improved and the performance of these algorithms is better than the algorithms not correcting NLOS error under different kinds of channel environmentFinally, location and tracking algorithms based on neural network are proposed. The algorithms are able to correct the NLOS errors using BP neural network on TDOA measurements.and then the positions of MS can be estimated by appropriate algorithm. Furthermore, cooperating correlation detection gate, the MS is tracked by the algorithm. The simulation results show that the algorithm performance is better than others not only in the static state but also in the dynamic state. Futher more, we proposed a new tracking algorithm in cellular networks. The algorithm using kalman filter is able to correct initial estimated location value of the MS. Furthermore, cooperating the detection gate, the MS is tracked by the algorithm. Simulations results show that the proposed algorithm could be efficiently track the MS, and has a good results.