Research On Key Technologies Of Indoor Localization

Lots of Location Based Services(LBS), such as the nearby food or hotel recommendation function included in some APPs have brought convenience to people’s life and inspired higher application requirement for localization at the same time. It is our ultimate goal to realize target localization of anytime, anywhere and with required accuracy. In open outdoor condition, satellite positioning has basically met the requirement, while no localization scheme has been extensively used in indoor environment so far.Indoor localization system can make use of two kinds of radio signals: signals designed for location determination such as UWB(Ultra Wide Band) signal, radio frequency signal, radar signal(pulse, continuous wave etc) and so on; signals transmitted for communication purposes such as cellular base station signal, WLAN(wireless local area network) signal and so on. This paper has chosen continuous wave signal and Wi Fi(Wireless Fidelity) signal as representatives of the two above mentioned signal forms respectively for further study based on existing indoor positioning algorithms.There are five innovation points in this paper, include:I. Proposed a localization algorithm suited for bistatic radar with continuous wave. The algorithm do not require synchronization between the transmitter and the receiver and it can obtain difference of distances between two arbitrary transceivers and the target. Combined with Kalman Filter, the algorithm can realize realtime localization of target in motion with changing speed. Simulation shows that given a random initial value the estimated position can converge to the true target location within a few iterations.II. Based on the direct correlation of RSSs(received signal strength) of the same AP(Access Point) collected at two different places, we came up with a new fingerprint matching algorithm in which the max correlation coefficient means the highest matching degree. Simulation and experimental results show that the proposed method has a better performance than the other two traditional schemes.III. Presented the likelihood function of Wi Fi RSS measurements under correlated shadow fading model. Deduced the theoretical lower bound of the estimation error of position estimation with Wi Fi RSS, i.e. the CRLB(Cramer-Rao Lower Bound).IV. Realized estimation of parameters in signal propagation model and proposed the maximum likelihood position estimation algorithm with Wi Fi RSS. Compared estimation error of the proposed scheme with its theoretical lower bound through simulation.V. Accomplished data collection in real laboratory environment. Conducted target localization with the reorganized experimental data and UJIIndoor dataset downloaded from the Internet respectively. Presented positioning performance of the algotithms in real-world scenarios.