The advent of GPS provided a new way of measuring surface displacements due to earthquakes by deploying GPS networks within active seismic areas. Japan is located in the confluence of several tectonic plates, hence its seismicity. In order to surveille this activity, one of wider GPS network in the world was deployed, i.e., GEONET. By processing data from 93 GEONET reference stations, we analyze the 2011 Tohoku-Oki earthquake using PPP strategy. We studied the time series during the event setting up a threshold value at we consider the time series are being altered by the earthquake. We also identified the time after the occurrence when the maximum displacements happen. With the study of these two parameters, we aim to show their different behavior as the main shock propagates along the Japan islands, with a focus on a better understanding of the earthquake and its propagation. To achieving this, a least square adjustment method was used to relate epicentral distance to topocentric displacements and the time of detection to epicentral distance. The results show an exponential behavior of the distance-displacement regression versus a linear behavior of the distance-time regression. Besides, we use the former linear regression to calculate and approximation of the velocity of the shock waves.