A Low-cost Fingerprint Positioning System in Cellular Networks

Abstract

An ultimate aim of mobile positioning research is to find a method providing high estimation accuracy to the user within minimum delay and at minimum cost. Conventional location techniques based on trilateration and triangulation rely on line-of-sight path between the base station antenna and the mobile unit. In densely built urban areas, this assumption is rarely valid. This fact degrades the location performance of the conventional techniques and motivates the need for development of more accurate technique suited for these areas. Positioning system developed in this research is divided into three sub-systems. The first sub-system solves the problems related to fingerprint localization and involves neural network as key element of the positioning algorithm. The postprocessing tasks which include tracking and map-matching are performed in the second and third sub-systems respectively.