Two forms of Friedland’s separate bias estimation algorithm with U-D factorization of the covariance matrices are provided. Each is suited to implementation in a particular computing environment. (We consider MATLAB and compiled computer languages.) We reduce the computation time substantially, primarily at the time propagation stage, by using a separated bias formulation, while retaining the desirable numerical properties of the U-D factorization. The perecentage reduction typically increases with ratio of bias state dimension to dynamic state dimension. A numerical evaluation is given for the MATLAB algorithm.