NO _x sensor-based state estimations for urea-based selective catalytic reduction (SCR) systems have attracted much attention in the past several years because of their significant importance in achieving high NO _x conversion efficiency and low ammonia slip as well as low operation cost. Most of the existing estimation techniques require sophisticated design processes and significant tuning effort, which may prevent them from widespread applications to production urea-SCR systems. In addition, the existing nonlinear dynamic observers may not be able to achieve accurate estimations fast enough due to the estimation error dynamics. The purpose of this study was to design a straightforward and effective NO _x sensor-based estimation algorithm for decoupling post-SCR NO _x sensor signals (NO _x concentration, ammonia concentration), and for estimating ammonia coverage ratio of urea-SCR systems. Singular-perturbation based approach was applied to decouple fast NO and NH ₃ concentration dynamics from slow ammonia coverage ratio dynamics. Two solution candidates can be obtained by utilizing fast SCR dynamic models and NO _x sensor model. Physics-based criteria were developed for further down selecting the solution candidates. Simulation verification results under US06 cycle proved the effectiveness of proposed method in accurately estimating the aforementioned key SCR states. The proposed estimation algorithm can potentially be popularly applied to production SCR systems for cost reduction and for further improvement of NO _x sensor-based advanced SCR control systems and on-board diagnostics.