Indoor Localization for Sparse Wireless Networks with Heterogeneous Information

Wireless indoor localization is crucial in ubiquitous computing environments. Although accurate and efficient indoor localization can be provided in dense wireless networks, most existing algorithms fail to locate a mobile user in sparse deployment networks. In order to address this issue, this paper presents a new fingerprinting localization algorithm based on cost function where received signal strengths from heterogeneous wireless networks are applied. To further improve the positioning accuracy, a spatial context constraint area for fingerprint matching is constructed based on the continuity and smoothness of pedestrian movement trajectories. Experimental results show that the proposed algorithm using heterogeneous information can significantly improve the accuracy of indoor pedestrian localization in sparse wireless networks.