TL-FCMA: Indoor Localization by Integrating Fuzzy Clustering with Transfer Learning

Abstract

With the rapid development of sensing, computation, wireless networking and communication technologies, many indoor location-based services have emerged, which promote technology of precise indoor localization in Internet of Things (IoT). Fingerprinting localization has become attractive because of its open access and low cost. However, wireless signals are easily affected by environmental conditions, which leads to localization models built from previous radio maps inaccurate. Reconstructing radio maps and updating them with fewer labeled new data while maintaining high-accuracy positions is a key but difficult problem. In this paper, an indoor localization algorithm integrating fuzzy clustering with transfer learning(TL-FCMA) is proposed. Fuzzy C-means clustering aims at minimizing the effect of regional environmental changes in the location area. With the result of fuzzy clustering, the radio map is reconstructed and the positioning is conducted with manifold alignment based transfer learning. Experimental results show that TL-FCMA has lower location errors than that of those compared algorithms.