Research On Visible Light Positioning Assisted By INS

As a new type of indoor positioning technology,visible light positioning measures the strength,angle,time difference of the received signal or information of the images of the light–emitting diode(LED)s to locate the terminal.The widespread use of LEDs in daily life reduced the cost of equipment of visible light positioning.In addition,for the reason of environmentally friendly and secure characteristics of visible light positioning,it has a great application prospect and has become a hot spot in indoor positioning.The paper focuses on the research of visible light positioning technology and the content included in the paper are as follows:1.Aiming at the problem of low positioning accuracy and localization probability when receiving few observations of LEDs,a visible light positioning algorithm based on the united observations is proposed.The proposed algorithm improves the positioning accuracy and probability of localization by increasing observations of LEDs in the position estimation model in the way of uniting the coordinates of images of LEDs and the received signal strength.The continuity of positioning results could be improved as well.To solve the united highly dimensional and nonlinear position estimation model,the unscented Kalman filter(UKF)is applied.Simulation results show that the proposed algorithm can effectively improve the positioning accuracy and probability of localization.2.Aiming at the problem of low positioning accuracy of pedestrian dead reckoning(PDR)algorithm of inertial navigational system(INS)leaded by low accuracy of frequency of steps,an iterative frequency estimation algorithm based on interpolated zoom spectrum is proposed.An unbiased frequency estimator based on complex signal model is deducted firstly.Then,based on the idea of two-steps estimation,the algorithm estimates the frequency of the signal by fast Fourier transform(FFT)and iteratively estimate does it again with the proposed estimator.Simulation results show high accuracy of estimated frequency of steps and low complexity of the proposed algorithm.3.To improve the continuity of positioning results,INS is applied to assist visible light positioning in the paper.Firstly,the reference measurement vector of the random attitude terminal is mapped into a unified coordinate system.The position estimation based on the mapped measurement vector removed the random error of coordinate system caused by the change of terminal attitude.Then,on the basis of the above analysis,the whole frame and the realization method of the visible light positioning based on INS are designed.Simulation results show improvements of continuity of positioning results of the proposed algorithm.Moreover,the proposed algorithm has higher positioning accuracy and smaller errors between the located trajectory and the real trajectory comparing with PDR algorithm of INS alone.