Research On Key Positioning Technology For Non-cooperative Node

The positioning of the non-cooperative nodes plays a vital role in many fields such as network reconnaissance and the environmental perception.Compared with cooperative positioning algorithms,there are more requirements and limitations for non-cooperative nodes positioning algorithm,which makes it more difficult.The positioning algorithm for non-cooperative nodes has been one of the fundamental steps as well as the key point in noncooperative nodes positioning.Among the existing researches,the estimations of Direction of Arrival(DOA)and Received Signal Strength Indicator(RSSI)can be used for positioning a single non-cooperative node.However,When there are multiple non-cooperative target nodes,the measurement information which belongs to one node needs to be correlated.The accuracy of traditional direction finding triangulation positioning algorithms and RSSIbased ranging positioning algorithms is susceptible to environment.These algorithms also fail to make full use of the position information contained in DOA and RSSI estimations.This thesis aims to study the positioning algorithms and data association algorithms for noncooperative nodes.On the premise that the signals of nodes can be intercepted,the DOA and RSSI of the incoming signals are measured.Then the measurements are used for calculating the target location.Specifically,based on the direction finding triangulation positioning algorithm,the fingerprint positioning algorithms and the S-dimensional assignment algorithm,the DOA-assisted fingerprint localization algorithm and location fingerprintassisted data association algorithm are proposed in this thesis.The RSSI data is collected under different scenarios for simulation experiments.The experimental results show that the positioning accuracy and the success rate of data association are effectively improved compared with the conventional algorithms.The main researches completed in this thesis are as follows:(1)Single node position reconnaissance: In this thesis,based on the researches of the rationale of spatial spectrum estimation algorithms,direction finding triangulation positioning algorithms and fingerprint positioning algorithms,the DOA-assisted fingerprint localization algorithm is studied.The RSSI fingerprints are normalized to deal with the changes in the transmitted signal strength of the target.In the online stage of fingerprint positioning algorithm,DOA information is used to construct the weight coefficients for the reference fingerprints which are selected from the fingerprint database.By taking the weighted sum of these reference fingerprints’ locations the final positioning results can be obtained.The results show that the positioning accuracy of the DOA-weighted fingerprint positioning algorithm is effectively improved compared with traditional algorithms.And the dependence of localization performance on parameters has also been improved.(2)Multiple nodes position reconnaissance: Traditional DOA data association algorithms are introduced in this thesis.Based on the multi-dimensional allocation algorithm,the location fingerprint-assisted data association algorithm is studied.When the position area is known,the match of fingerprints narrows down the possible area of the target nodes,which can reduce the computational complexity of data association process.The simulation results show that the assistance of location fingerprint can effectively improve the success rate of data association.