This paper addresses a problem of observer-based sensor fault reconstruction for continuous-time systems subject to sensor faults and measurement disturbances via a descriptor system approach. An augmented descriptor plant is first formulated,by assembling measurement disturbances and sensor faults into an auxiliary state vector. Then a novel descriptor state observer for the augmented plant is constructed such that simultaneous reconstruction of original system states, sensor faults and measurement disturbances are obtained readily. Sufficient conditions for the existence of the proposed observer are explicitly provided, and the application scope of the observer is further discussed. In addition,an extension of the proposed linear approach to a class of nonlinear systems with Lipschitz constraints is investigated. Finally, two numerical examples are simulated to illustrate the effectiveness of the proposed fault-reconstructing approaches. This paper addresses a problem of observer-based sensor fault reconstruction for continuous-time systems subject to sensor faults and measurement disturbances via a descriptor system approach. An augmented descriptor plant is first formulated, by assembling measurement disturbances and sensor faults into an auxiliary state vector . Then a novel descriptor state observer for the augmented plant is constructed such that simultaneous reconstruction of the original system states, sensor faults and measurement disturbances are obtained readily. Sufficient conditions for the existence of the proposed observer are explicitly provided, and the application scope of the In addition, an extension of the proposed linear approach to a class of nonlinear systems with Lipschitz constraints is investigated. Finally, two numerical examples are simulated to illustrate the effectiveness of the proposed fault-reconstructing approaches.