High ground watertable is one of the main problems that can disrupt the tunnel construction. The prediction of groundwater infiltration into the tunnel is very important subject for preventing problems during drilling and designing drainage system. Therefore, it is necessary to consider the amount of water flow and drainage model in the design, thus, it is necessary to forecast quantity of water flow and the drainage model and consider them in design stage. For this purpose, to present a dewatering design in the eastern part of Tehran Metro tunnel Line 7, the cross-section of 4+ 500 + 4 +600 Kilometers , two drainage galleries with a length of 14 meters were used. Modeling and estimating the entering water into the tunnel requires detailed and accurate information about soil hydrodynamic properties. Also, the accuracy of modeling and numerical analysis and comparison with actual conditions for optimization of the design is very important. One of the best methods for evaluating these parameters in ground mass scale is back analysis according to experimental results. Ground water infiltration into two galleries was calculated for the whole length of the tunnel in steady state, using from finite element and finite difference methods in PLAXIS® and FLAC® softwares.Then, from the data obtained from the Lefranc test, the value of the soil hydrodynamic parameters were estimated by the method of back analysis. The obtained results indicate the value of the method of return analysis compared to the Levfran method for measuring permeability with the same data. The measured value of groundwater infiltration in field tests was 600 cubic of meters per day and the error of the finite element and finite difference methods is equal to 12 and 4/8 percent, respectively, which are in a good agreement with the field measurements. The optimum value of permeability parameter obtained from back analysis equal to 4.5  10 -5 and 3.6  10 -5 meters per second was set for this area as well as optimum transmissivity equal to 117 and 214 meters per day.