Mobile terminal location estimation has attracted much interest for its applications in emergency calls, intelligent traffic control, intelligent transportation systems and resource allocation. In this paper, a new implementation of data fusion technique for discrete mobile terminal location which is based on Support Vector Machine (SVM) is introduced. The raw information, i.e., time of arrival (TOA), angle of arrival (AOA) and received signal strength (RSS) obtained from the survey points in a real dense urban environment, is post processed to create feature space. The feature space, which contains the information on the locations of the mobile users, is classified by a previously trained SVM network. The simulation results demonstrate that the location estimation error is less than 100m in almost 72% of the calls.

Keywords