Location Determination with Factor Graphs for TDoA

This paper covers time difference of arrival (TDoA) based positioning in wireless cellular networks in a factor graphs framework. The occurring non-linear estimation problem of the localization process cannot be solved analytically. One well-known solution is based on the Gauss-Newton algorithm. But this iterative standard approach fails to converge for certain geometric constellations and bad initial values, and thus, it is not suitable for a general processing in cellular networks. Therefore, we introduce a more robust TDoA positioning algorithm using the method of factor graphs. Simulation results in terms of cumulative density functions show the superiority of this approach compared to Gauss-Newton