Hybrid indoor localization method based on signal subspace fingerprint database

Abstract

To solve the localization problem of multipath propagation in complex indoor circumstance, a localization method of signal subspace matching based on fingerprint database is proposed by using small antenna array in the indoor environment. Compared to the RSSI, the signal subspace fingerprint can obtain better effect by utilizing more space information. The received signal from each array is firstly processed with self-correlation and it's eigenvalue decomposed to create the signal subspace fingerprint. Location is then determined by the smallest angle between the received signal subspace and the fingerprint database. Simulation results show that the proposed algorithm has made a great improvement in localization accuracy.