In this paper, we introduce a signal transformation (ST) methodology for tracking control of a large class of arbitrary references. The method can improve tracking performance of ordinary one-degree-of-freedom (1-DoF) feedback control structure, while keeping robustness against unmodeled dynamics and limiting the projected measurement noise by ensuring a low closed-loop bandwidth. Using singular perturbation theory, sufficient conditions for stability and convergence of the tracking error are derived. Effectiveness of the proposed method is demonstrated by simulations. It is shown how ST method can provide a better control performance compared to ordinary 2-DoF feedback control systems having similar projected noise power, and maintain robustness against uncertainties, disturbances, and unmodeled dynamics.