蜂窝无线网络中移动台定位算法Fingerprint的研究

In the study of positioning algorithm in wireless cellular network, we found that the CELL-ID algorithm has a relatively low accuracy, and the accuracy of positioning algorithm based on time of arrival and time difference of arrival is easily been influenced by the measuring distance; GPS has a better accuracy, but it also has a high demand on the terminals, and it is eaily influence by the weather and environment. So we study the Fingerprint algorithm, this algorithm has a lower deviation by the NLOS propagation, and it is relatively easily to be achieved when there is a certain amount of measured data. In this paper, we use the idea of Fingerprint algorithm and the SAS tool, with the base of road test data. Firstly establish the Fingerprint database, then use the cluster analysis and discrim analysis in SAS system, from two aspects to verificate: based on the longitude and latitude, and based on BCCHLevel. And the result shows that the accuracy of Fingerprint algorithm is better than that of other traditional positioning methods.