Research On Visible Light Compressed Sensing Positioning Algorithm In Complex Indoor Environment

Nowadays,with the rapid urbanization,indoor space is becoming more and more complex.Under the increasingly urgent demand,indoor positioning has attracted people's attention.With the rapid development of LED lighting technology,visible light communication technology has been developing rapidly.And compared with the ratio frequency communication,the visible light communication has many significant advantages.So indoor positioning based on visible light communication has gradually become a research hotspot.In this paper,the visible light localization system is analyzed firstly,and the direct channel model of indoor visible light localization is introduced.On the basis of the ideal direct channel model,the noise model is established as the interference factors in the indoor positioning system.Considering the reflected light caused by unavoidable opaque objects in the actual positioning,a reflection model which is more consistent with the actual positioning environment is obtained by combining the diffuse reflection with the specular reflection in a certain proportion.Several indoor visible light positioning algorithms are analyzed,including the Received Signal Strength(RSS),Time Difference of Arrival(TDOA),Angle of Arrival(AOA).The fundamental positioning principle of the various algorithms is given.In addition,the multilateral measurement algorithm based on RSS is introduced in detail.With the purpose of reducing the influence of environment factors such as noise and reflected light,a novel visible light indoor positioning algorithm based on compressive sensing is proposed.In this algorithm,the target position was defined as a sparse vector in discrete space,and the power measurement matrix received by the target is expressed as the product of the measurement matrix,the sparse matrix and the sparse vector in the compressive sensing theory,and the sparse signal reconstruction algorithm is used to recover the target position,which can effectively reduce the impact of indoor environment on positioning accuracy.In order to reduce the positioning error due to occlusion,a pre-processing procedure is introduced.When the target is not in the center of the grid,but near the boundary of the grid,the compressive sensing algorithm may locate the target in the nearby gi'id in the case of environmental factors,and the positioning result has large positioning error.Therefore,a fusion positioning algorithm using compressive sensing and multilateral measurement is proposed.Firstly,compressive sensing algorithm is used in the coarse positioning stage to judge the position of the grid where the target is located.If the target is not within the recovery range,we adopt multilateral measurement algorithm for accurate positioning of the target.In the application of multilateral measurement algorithm,the base station selection strategy is used to optimize the fusion algorithm,which improves the adaptability of the fusion algorithm to indoor complex environment.