Abstract: Aiming at improving both performances including accuracy and real time on the application of Wireless Sensor Networks (WSN) user location estimation, a localization approach that employs Gauss Mixture Model (GMM) to analyze received signal strength is proposed in this study. This approach consists of an offline training phase and a real time localization phase. In the offline phase the signals came from many training locations were recorded and the GMM parameters were calculated and in the real time phase this GMM was employed to match the received signal strength patterns against the training patterns. The signal strengths were modeled by GMM can improve model precision in the training phase and further increased accuracy of location estimation in the localization phase. Experiment results demonstrated that the proposed method had better performances on both accuracy and real time of WSN user localization.