Research On Fingerprint Positioning In Wireless Networks

With the rapid development of wireless communication technology, a wide varity of wireless networks have emerged. Only a single kind of wireless network can not meet the requirement of wide coverage, low latency, high bandwidth and low cost, therefore the heterogeneous network has become an inevitable trend of wireless network.The realization of many value-added services in homogeneous network must relay on positioning technology which is also very important in heterogeneous network.Research and analysis on TDOA、AOA or other traditional positioning technologies is based on a single network, and these positioning technologies are more vulnerable to the impact of NLOS propagation and multiple propagation. How to overcome the influence of NLOS propagation and look for an simple and high precision position algorithm have a certain theoretical and practical significance to speed up the development of positioning technology and its application in engineering.Because of the good ability to overcome the effct of NLOS propagation,the fingerprinting positioning has become one of the hot topics of wireless location technology in recent years.The thesis is based on DCM(Database Correlation Method)fingerprinting positioning technology,the existing network structure is analyzed and improved,and the positioning performances of the relevant algorithms in WLAN、WCDMA and heterogeneous network are reserched with an emphasis.Firstly, the present research status of fingerprinting positioning technology is analyzed, and the limitations and shortcomings of existing postioning technology are summarized based on the access to lots of literatures.Secondly, the principles of the existing positioning algorithm and the evaluation criteria of the positioning accuracy are briefly introduced, the location-enabled WCDMA and WLAN network structure and heterogeneous network integration architecture is overviewed. Combined with the characteristics of the fingerprinting position algorithm, the existing network structure are analyzed, explored and improved in this thesis, and a location-enabled heterogeneous network structure is designed.Again, based on the detailed theoretical analysis on the DCM algorithm and SVR algorithm under WCDMA environment, the improved algorithm based on Cell ID is proposed in this thesis. The simulation results showed that the improved algorithm has a higher positioning accuracy. Then the DCM algorithm and SVC algorithm are analyzed and an improved positioning algorithm is proposed and compared, the simulation comparations have indicated the effectiveness of the improved algorithms. Then, the effect to the positioning accuracy of the AP number and the BS number are analyzed based on the front positioning simulation. Combined with the characteristics of the location fingerprint positioning technology and the best linear fusion algorithm,two ideas and methods of data source integration and decision-making fusion are made.The positioning performances of fusion positioning algorithm are compared and analyzed under different scenarios,and the positioning algorithms which should be used in the different networks of heterogeneous network environment are summaried.Finally, the research work carried out in this thesis is summarized and the contents and recommendations for further research are given.