Research On Wireless Localization Algorithm Based On BP Neural Network

With the rapid development of wireless communication technology, wireless location technology is playing the more and more major role in People’s Daily life, obtaining the position information of the mobile terminal rapidly and accurately and providing the demand of location service are becoming more and more urgent. Therefore, the localization technologys for obtaining position information of mobile terminal quickly and accurately and the positioning systems have become a hot spot in the research in different kinds of wireless network systems.In reality, using one positioning technology for mobile station location, often due to the limitations of the positioning technology, lead to can not realize positioning or positioning effect is not very ideal. With in-depth study of wireless location technology, mixing different localization technologies, make up for their respective defects, and play their respective advantages, the development trend of wireless location is to realize seamless and high precision localization.First of all, a hybrid localization algorithm based on BP neural network is proposed in this paper, when base station or GPS satellite positioning independently, in order to solve the problem of difficult to locate because of the mobile terminal can not observe enough number of GPS satellite or measurement data of base station location is insufficient. Firstly, the base station TDOA measurements and GPS pseudorange difference are modified by using BP neural network, then the position of the target mobile station is estimated by using location algorithm. Finally, the proposed location algorithm is simulated and evaluated in the non-line-of-sight environment. The simulation results show that the proposed algorithm can realize the localization for mobile station, the location error is obviously lower than the other location algorithm, also show that the influence of non-line-of-sight error is effectively eliminated by the proposed location algorithm, and the location accuracy is improved.With the widely application of the mobile station location, the real-time locating and tracking service for mobile station are need to provide while the mobile station is in a state of motion, a markov chain location and tracking algorithm based on BP neural network is proposed. Firstly, in order to eliminate the influence of non-line-of-sight error, the measured datas are modified by using BP neural network. Then the position of mobile station is estimated by using the hybrid location algorithm. Finally, the tracking for mobile station is realized by using the markov chain cooperate with relevant detection gate. Simulation results show that the proposed tracking algorithm can realize the dynamic tracking for mobile station, and have good tracking performance.Then, a novel kalman filter tracking algorithm based on BP neural network is proposed. Firstly, BP neural network is used to correct the non-line-of-sight error exist in the locate measurement datas. Secondly, the initial position of the mobile station is estimated by the hybrid localization algorithm. Then, the initial position of the mobile station is corrected by the kalman filter, the tracking for mobile station is relized cooperate with relevant detection gate. The effectiveness of the proposed algorithm is demonstratey by the simulation results.Finally, the detailed summary is done for the whole text content, some problems of the research involved in the mobile station location are discussed, and the development of wireless location technology is prospected.