Research And Application On Riadio Positioning Technology In LTE Network

LTE network is proceeding quickly all over the world in recent years. LTE industry chain is more and more mature. LTE network has established a dominant position on the next-generation mobile communications market. The development and deployment of LTE put forward higher required to Location-Based Service(LBS) and presented more challenges and required to position testing technology.This paper firstly give the principles and characteristics of mobile-based location, network-based location and GPS assisted location. And the pager give introduction of basic positioning algorithms such as time of arrival TO A, time difference of arrival TDOA, Angle of Arrival AOA based positioning algorithms and hybrid positioning methods. The factors affecting accuracy and error are introduced later. Several methods for the identification and elimination of NLOS errors are also introduced. This paper described the LTE positioning technology and explained fingerprint location method, a supplement LTE position method.This paper rely on these theories and knowledge. With the help of SAS statistical analysis tool, the paper process the cluster analysis and discriminant analysis using fingerprint method. Details are as following:(1) Analysis of the existing network data, filter out the data which meet the test requirements and with no obvious erroneous from existing network data.(2) Design two experiment case, fingerprint positioning based on latitude and longitude and fingerprint positioning based BCCHLevel. Starting from the characteristics of the two cases, I screened sample data from the raw data. With the help of cluster analysis tool in SAS, I classified the samples according to different cluster standard, and then established the fingerprint data map.(3) I extracted the positioning data from raw data, and imported these data into SAS to proceed discriminant analysis. And I design and write the positioning discriminant code. Then I discriminate and analysis the positioning data, draw the scatterplot map and CDF map according to the discriminant result. At last I analysis the algorithm performance of fingerprint such as precision under different cases.