Projected capacitive touch sensing is the first choice in smartphone and tablet market,and noise mitigation is the main obstacle in product development.Traditional methods focus on how to measure the value of mutual capacitor accurately under noisy environment.This paper presents an algorithm which uses kalman filter for noise mitigation in touch systems.The process noise covariance and measurement noise covariance change and are updated with each measurement.The filter estimates and predicts the behavior of the next output based on the dynamic model of touch event and then uses the measured value to try to correct it.The experimental results show that out approach can improve the touch performance even under the noisy environment.