Research On Indoor Multi-Node Cooperative Positioning Algorithm With NLOS Information

With the widely application of Location Based Services(LBS)in military and civil fields,indoor and outdoor positioning technologies have become hot research issues at present.In the outdoor environment,the Global Navigation Satellite System(GNSS)can provide accurate location information.However,due to the obstruction of walls and other obstacles in the indoor environment,effective location estimation cannot be provided.Usually,there are many obstacles in the indoor environment,so that part of the signal transmission environment is Line of Sight(LOS)environment,and part of the signal transmission environment is None Line of Sight(NLOS)environment.When the signal propagates in the NLOS environment,there will be a relatively large NLOS noise in the measured distance which is the main factor affecting indoor positioning accuracy.Therefore,it is the main research objective of this paper on how to reduce the impact of NLOS noise on indoor positioning accuracy.In order to solve the above problems,this paper first proposes a reverse positioning method based on trilateration positioning.Two virtual anchor nodes are added to help the target node’s estimated location reverse locate the original anchor nodes’ location.The error of the estimated location of the target node is judged according to the error of the anchor node in reverse positioning.Through theoretical analysis and simulation experiments,it can be obtained that the greater the noise contained in the measured distance,the greater the estimated position error of the target node,and the greater the anchor node error of reverse positioning.Based on the reverse positioning method,this paper has introduced the cooperative positioning into the indoor positioning system,and then gives two positioning algorithms aiming at the indoor multi-nodes cooperative positioning system containing NLOS information.The proposed algorithms are as follows:The first one is an indoor positioning algorithm aiming at indoor environment with unknown NLOS information.Firstly,an NLOS identification algorithm is proposed to determine whether there is NLOS noise between the target node and the three anchor nodes according to the error of the anchor nodes in reverse positioning.Afterwards,based on NLOS identification algorithm,a new NLOS-LS algorithm is proposed.The algorithm helps each target node to select three anchor nodes without NLOS noise for positioning through a two-round positioning process,so as to reduce the influence of NLOS noise on the positioning accuracy of the target node.This algorithm does not need any prior information about NLOS noise to identify NLOS.It can effectively improve the target positioning accuracy when the indoor positioning environment with a small amount of NLOS information.Simulation experiments show that under different NLOS noise distributions,the identification success rate of the NLOS identification algorithm can reach more than 95%.Its positioning accuracy is obviously better than other contrastive algorithms and the positioning results are stable.The second one is an indoor positioning algorithm aiming at the indoor environment with known NLOS information.In the case that the NLOS information between nodes has been identified,an iterative NLOS error mitigation algorithm,EM-LS algorithm,is proposed.The algorithm introduces an error mitigation value,which increases in the iteration process.That subtracts the above value from the distance containing NLOS noise,and then reverse locate the anchor node.The termination condition of the iteration is that the error of the anchor node in reverse positioning does not increase.This algorithm does not need any prior information about NLOS noise to estimate the size of NLOS noise,and the error is subtracted from the measured distance to reduce the impact of NLOS noise on positioning accuracy.Simulation results show that the EM-LS algorithm has higher positioning accuracy and stable positioning results under different NLOS noise distributions.