A Bayesian Method to Improve Mobile Geolocation Accuracy

This paper presents a geolocation method for GSM/CDMA/UMTS networks. A call area is partitioned into a grid and a sequential Bayesian updating scheme is proposed to identify the grid point within the circular belt, defined by the one-way delay between the mobile and the base station, that is the most likely position of the user. The one-way delay value is characterized by Timing Advance (TA) in GSM, the Round Trip Delay (RTD) in CDMA, and the Round Trip Time (RTT) in UMTS. Using GSM as case study, we employ an RF simulation model to study the accuracy of the Bayesian update method is relatively insensitive to cell size and fairly robust to the parameter setting. It can offer performance superior to existing methods that are not GPS assisted in rural and surburn=ban environments.