Large-scale Wireless Fingerprints Prediction for Cellular Network Positioning

Cellular network positioning is a mandatory requirement for localizing emergency callers, such as E911 in North America. Although smartphones are normally with GPS modules, there are still a large number of users with cell phones only as basic devices, and GPS could be ineffective in urban canyon environments. To this end, the fingerprinting positioning mechanism is incorporated into LTE architecture by 3GPP, where the major challenge is to collect geo-tagged wireless fingerprints in vast areas. This paper proposes to utilize the subspace identification approach for large-scale wireless fingerprints prediction. We formulate the problem into the problem of finding the optimal subspace over Stiefel manifold, and redesign the Stiefel-manifold optimization method with fast convergence rate. Moreover, we propose a sliding window mechanism for the practical large-scale fingerprints prediction scenario, where fingerprints are unevenly distributed in the vast area. Combining the two proposed mechanisms enables an efficient method of large-scale fingerprints prediction in the city level. Further, we validate our theoretical analysis and proposed mechanisms by conducting experiments with real mobile data, which shows that the resulted localization accuracy and reliability with our predicted fingerprints exceed the requirement of E911.