Research On Anchor Node Position Optimization And Multi-node Positioning For Indoor Environment

There are usually many obstacles in the indoor environment,such as tables,pedestrians,iron doors,concrete walls,etc.,which will form signal propagation under non-line-of-sight conditions during signal transmission,thus affecting the transmission and reception of signals.If there is a large amount of non-line-ofsight information or information loss between the target node and the anchor node,it will cause the positioning accuracy of the target node to be reduced or cannot be located.In the indoor environment,this situation is more common,but there may be other known nodes to be located within the communication range of the node,and through communication with neighboring nodes,a multi-node positioning network is formed,useful location information or communication messages are obtained from neighboring nodes,or indirect use of neighbor information for positioning.Therefore,how to design high-precision,low-complexity distributed collaborative positioning algorithms is an important challenge in indoor multi node positioning networks.Currently,the mainstream distributed collaborative positioning schemes based on ranging have three important issues:(1)Due to obstacles blocking the ranging equipment,there are data with large ranging errors in the received data,which seriously affects the final positioning results;(2)When the distance between two node sensors cannot be measured,the Euclidean distance is replaced by the addition of multiple hop distances from multiple sensors that can be ranging in the middle,resulting in poor positioning accuracy after solution;(3)The clustering method used in distributed algorithms is unreasonable,resulting in complex computing or resource waste.In order to solve these problems,this paper studies the related technologies of anchor node location optimization and multi node distributed positioning for indoor complex environments.The main work is summarized as follows:Firstly,in view of the limited distribution of anchor nodes in the current indoor scene,a method of synchronous calibration and positioning of anchor nodes and target nodes based on distance measurement is proposed to fuse IMU sensor data for mobile nodes.Firstly,data fusion is performed between the IMU sensor carried by the mobile node and the distance measurement data of the anchor node,and the relative position of the movement trajectory relative to the known anchor node is obtained,Then,the distance measurement data of the virtual node formed when moving the node to the unknown anchor node is used to reverse locate the unknown anchor node.Then,aiming at the problem of poor positioning accuracy of multiple nodes in indoor positioning environments,a rigid wheel graph clustering Wheel-MDS(P)algorithm based on UWB distance measurement is proposed.Firstly,the measured data obtained by the system are processed,and the edges that do not meet the constraint conditions are counted and eliminated through trilateral constraints.Then,a gradient descent method is used to complete the missing data in the Euclidean distance matrix,effectively reduce the impact of missing error data on the overall positioning accuracy.Next,based on the wheel graph rigidity principle,the multi node positioning network is clustered to reduce the system’s computational complexity.Then,the final coordinate position of the system is obtained by using a common node fusion method and matrix transformation method between clusters,improving the overall positioning accuracy of the system and reducing computational complexity.Through simulation experiments,the optimization of data elimination and matrix completion algorithms on the overall data is verified.The experiments verify the performance improvement of Wheel-MDS(P)algorithm on MDS-MAP(P)and MDS-MAP algorithms.