A metric learning model for localization

With the rapid development of intelligent mobile devices, cell phone localization has become an important research problem. Fingerprinting based methods have been widely used to address this problem, in this paper, we propose a fingerprinting based method by using distance metric learning. We present the details of the method and show how to learn a Mahalanobis distance metric for localization. To evaluate the performance of the method, we compare it with maximum posterior probability method and the traditional K nearest neighbors method. Results show that the proposed method can significantly improve the accuracy of the localization in urban areas.