Fingerprint- And kalman filter-based localization exploiting reference signal received power calibration

This paper proposes a localization scheme exploiting reference signal received power (RSRP) for estimation of the next location. The proposed scheme can correct outliers without discarding data by adding RSRP as a state vector for a Kalman filter, and combining the Kalman filter with fingerprint-based localization. Performance evaluation is carried out via simulations in indoor environments. Results indicate that the proposed scheme can effectively correct outliers and enhance positioning accuracy. The root mean square error in the positioning error was reduced by 56%, compared to the conventional fingerprint-based localization schemes for indoor environments.