This paper focuses on the problem of sensor fault estimation of switched fuzzy systems with unknown input. By treating the sensor fault as a virtual state, an augmented system can be obtained. A dynamic unknown input observer is designed for the augmented system to estimate the sensor fault and the system state simultaneously. Based on the average dwell time and the piecewise Lyapunov function, the observer parameter calculation method is presented in terms of solutions to a set of linear matrix inequalities. The unknown input is decoupled from the estimation error, and the effect of the disturbance is attenuated by the weighted H_∞ performance level. At last, simulation results are provided to illustrate the effectiveness of the proposed techniques.