The goal of this research is to develop both nonlinear and linear state estimation strategies for both SCR and DPF devices. Development of "device health" outputs, as a function of real-time estimated states and nominal-operation states is an objective. Proving stability and quantifying the accuracy of these methods will be a primary objective. Finally, developing rational metrics for comparing trade-offs of computational complexity and accuracy will be developed. The objective of this research was to develop estimators for DPF and SCR state estimation. The SCR state estimator is based on the high fidelity model in reference [4, 5]. The DPF state estimator is based on the high fidelity model in reference [6] SCR High Fidelity Model Development A high fidelity multi-step global kinetic SCR model is important for studying the kinetics of the SCR reactions and provides a basis for developing the SCR reduced order model and state estimator. The performance of the SCR models and state estimator are highly dependent on the kinetic model parameters. SCR reactor and engine experimental studies and model calibration to experimental data are necessary for developing the SCR high fidelity model and identifying the model parameters. The high fidelity SCR model was developed from an existing model calibrated to Fe-zeolite reactor data [7]. The model was modified for Cu-zeolite SCR by changing the mass transfer correlations and adding necessary reactions. The model considers of a 1-D single channel of the SCR catalyst which was axially divided into a series of axial elements. The model contains equations describing the mass transfer, heat transfer and a series of SCR chemical reactions. It simulates the species concentrations in both the gas phase and surface phase for NO, NO 2 , NH 3 and N 2 O gas and wall temperatures, as well as the NH 3 storage along the axial distance of the channel.