Research On Key Technique Of Distributed Measurement And Positioning

Compared with Wireless Sensors Networks(WSN),Mobile Ad Hoc Network(MANET)is a typical distributed network with large spatial scale and high dynamics,which is widely used in manned / unmanned spaces Scenarios such as collaboration,tactical vehicle networking,and aircraft clusters carry systematic functions such as collaborative formation,collaborative detection,collaborative confrontation,and collaborative control.Such systematic network external function bearer strongly depends on the relative position awareness and network topology control of network nodes based on relative measurement positioning within the distributed collaborative network.In response to this problem,this paper studies the key technologies of distributed joint measurement positioning.The main research contents and innovation contributions include:1.Propose an integrated communication measurement mechanism based on distributed network.The main features include:(1)Based on the existing distributed communication architecture,add time information,transmitted signal strength information,cooperative relay hop information and other element identification for relative measurement and positioning in the information reserved bits of the transmission data frame identification field;(2)Extract the corresponding element identification information through the frame analysis process of the receiving unit: use the logic resources of baseband signal processing to implement the core algorithm of relative positioning of network nodes.(3)The proposed integrated measurement and measurement mechanism does not change the existing distributed network communication architecture,and utilizes the network's regular control information interaction and data interaction to achieve relative measurement positioning of network nodes.This mechanism does not require additional wireless transceiver devices or links,and does not occupy additional network physical resources.2.A 3-D clustering MDS relative positioning algorithm based on MANET systemization is proposed.The algorithm has the following characteristics:(1)Based on MANET's distributed network topology control principle,according to the order of network node connectivity,based on the optimal criterion of connectivity,select the cluster head to achieve network clustering and spatial clustering;(2)Based on the basic principle of Multidimensional Scaling(MDS),the geometric center of the polyhedron constructed by all nodes in the cluster is used as the origin of each local coordinate system;three straight lines passing through the origin orthogonally are selected as the local relative coordinate system.Standard direction,establish a local coordinate system,calculate relative coordinates according to the projection of each straight line in the unit direction,and realize local positioning of the distributed network;(3)Based on the number of neighbor nodes and network connectivity,the nodes are networked,the network is divided into first-level core clusters and second-level general clusters,and the second-level general clusters are classified according to the origin and standard direction of the local coordinate system The positioning results are translated and rotated.The coordinate parameters of the same cluster head node in the corresponding local coordinate system are used to solve the corresponding translation and rotation parameters of each cluster to realize the fusion of the local coordinate system to the core coordinate system.Compared with the traditional "pass-through" distributed positioning algorithm,this algorithm relies on the characteristics of dynamic networking and multiple transmissions of distributed networks,and clusters and clusters network nodes according to the characteristics of network space distribution to achieve the "Parallel operation",each cluster performs relative positioning of network nodes synchronously,reduces the algorithm convergence time,reduces the cumulative error caused by positioning transfer,and improves the network measurement positioning accuracy;The numerical simulation results show that compared with the traditional AHLos algorithm,the proposed MDS relative positioning algorithm based on MANET 3-dimensional clustering has higher positioning accuracy and faster convergence time.When the number of network nodes is 32,after 5 iterations,the proposed algorithm converges,the average positioning error is 2.984,and the pure average average positioning accuracy is 4.564%.Compared with the AHLos algorithm,the average positioning error is improved by 53.846%;the cumulative error of the edge nodes is reduced by 56.632%.The distributed relative measurement and positioning technology studied in this paper can be applied to network node perception and control in a systematic collaborative scenario.It has important theories and realities for the optimization of group functions of network node cooperation,topology construction and reconstruction,and network dynamic routing.significance.