Instead of using conventional template matching with the sum of squared distance (SSD) for signal strength matching to provide user location on wireless infrastructure [Radar: an in-building rf based user location and tracking system], we propose the use of robust template matching methods [Fast Algorithm for Robust Template Matching with M-estimators] for providing robust user location. Our method expects and handles the noisy data common within a wireless setting, and provides the user with relevant information as to the reliability of their location within the system. In our experiments, we compare our method with conventional SSD based and sum of absolute difference (SAD) based template matching and a state of the art method, support vector machines [The Nature of Statistical Learning Theory]. Our experimental results show that some of robust template matching methods can outperform compared methods very much.