Research On Indoor Positioning Algorithm Of Single LED Lamp Based On A-Bayes

Visible light indoor positioning technology is a kind of indoor positioning technology which uses light-emitting diode(LED)as transmitter and photodetector(PD)as receiver.This technology is popular because of its high positioning accuracy,low equipment complexity and low equipment cost.At present,the mainstream visible light indoor positioning technology is still using multiple LED lights for positioning,and in the multiple LED lights indoor positioning technology,how to distinguish light sources and how to reduce the interference between light sources has become a difficult problem.In some indoor places,the sparse layout of light source will lead to the signal incidence angle of a light source will be greater than the field of view angle of PD,and the positioning error is large.In some small space occasions,the use of multi LED indoor positioning will lead to a waste of light source and additional cost.The indoor positioning with single LED lamp can effectively solve these problems in the indoor positioning technology of multi LED lamps.Therefore,this paper aims to seek efficient indoor positioning technology of single LED lamp,improve the traditional visible light indoor positioning algorithm,and solve the problems of indoor positioning technology of multi LED lamps.This paper proposes a new four-PD single-lamp positioning model.The transmitter uses a single LED lamp,and the receiver consists of three inclined PDs and one horizontal PD.Based on the new four PD single lamp location model,two location algorithms are proposed.Firstly,a high-precision indoor positioning algorithm based on the received optical power ratio is proposed.The three-dimensional coordinates of the receiver are obtained by using the received optical power ratio of tilted PD and horizontal PD,as well as the tilt angle and azimuth angle of each PD.The algorithm is simulated and experimented in the range of 3m×3m×3m,the simulation results show that the average positioning error is 13.55cm,the points with positioning error within 15cm account for70% of the total test points,and the points with positioning error within 20cm account for more than 90% of the total test points.The experimental results show that the average positioning error is 1.23m at 0m height.Then a high-precision indoor location algorithm of single LED lamp based on a-bayes is proposed.On the basis of four PD single lamp location model,machine learning algorithm is introduced.AdaBoost strong classifier is used to divide the test area into center and edge areas.Bayesian matching algorithm is used to match the areas.Simulation and experiment are carried out in the range of 3m×3m×3m,and better location effect is achieved.The simulation results show that the average positioning error of the algorithm is 4.38cm in the range of 3m×3m×3m,and the test points with positioning error within 5cm account for more than 62% of the total test points.The experimental results show that the average positioning error is 9.27cm,and more than80% of the total testing points have positioning error less than 15cm.The new four PD single lamp positioning model proposed in this paper is more novel than the traditional visible light indoor positioning system model.The receiver is small and light,several PD can get different received optical power values in a very short distance,and the construction of fingerprint library is more convenient.The two indoor positioning algorithms of single LED lamps have high positioning accuracy,which do not need to consider the interference between light sources,do not need to distinguish light sources,and the equipment is simple,which are suitable for the situation of sparse layout of light sources and small space scope,which can effectively solve the problems existing in traditional indoor positioning technology of multi LED lamps.