Localizing the mobile robot in an indoor environment is one of the problems encountered repeatedly. Achieving the target precisely in any environment is not an easy task since there are noises and obstacles in the surrounding environment. Therefore, filtering the signals to reduce noises is essential for more accurate and precise motion. In this paper, we selected the extended Kalman filter, which is used for non-linear models' signals to predict the coordinates of a wheeled mobile robot. We tested the efficiency of this filter under three noise cases: no noise, Gaussian noise and non-Gaussian noise using MATLAB software.