RSS-based SVR models for GSM and DCS mobile users´ localization

A big interest in developing commercial LBS (Location Based Services), and also the obligation of mobile network operators to implement emergency call service, have led to the intensive development of techniques for mobile users localization in the last few years. In this paper, three different types of positioning models, based upon utilization of RSS (Received Signal Strength) values from GSM (Global System for Mobile Communications), DCS (Digital Cellular Service) or both GSM and DCS base stations, are proposed. Models were implemented using Support Vector Regression algorithm. Training and verification of proposed models was done by using real data, measured in a suburban environment. The accuracy of the proposed positioning models was compared and analyzed in terms of the number of radio visible base stations at measured locations.