Research On High-precision Ranging And Time Synchronization Technology Between Nodes In Ad Hoc Network

Ad Hoc is a network architecture that does not require specific infrastructure,adopts distributed control,and has the characteristics of self-organization,no central node,and multi-hop.In some modern battlefields,emergency rescue and other occasions play an active role and have broad development prospects.With the continuous development of self-organizing network technology,higher requirements are put on the ranging system of the Ad Hoc network.In the Ad Hoc network,time synchronization is one of the key technologies that need to be researched.The functions of self-organizing network nodes such as autonomous positioning and autonomous timing all depend strictly on precise time synchronization between nodes.Based on the problems of high-precision ranging and time synchronization between nodes,this thesis mainly does the following work:(1)A channel modeling method suitable for ranging technology research is proposed.Modeling multiple channel models for Gaussian channels,multipath channels,and interfering channels.By comparing the RMSE curves of the ranging accuracy in different channel environments,the correctness of establishing the channel model is illustrated.A method for identifying the best ranging path based on fuzzy least squares is proposed.Through simulation verification,the effectiveness of this method is proved,which can overcome the influence of harsh NLOS multipath channels and achieve the goal of high precision ranging.Aiming at the problem of large path recognition error in the case of low signal-to-noise ratio,this thesis proposes a noise reduction method for ranging signals based on Kalman filtering method.Through simulation verification,the results show that under low SNR environment,the noise reduction method of the ranging signal based on Kalman filtering can effectively reduce the noise of the received signal.(2)The node clock model of the Ad Hoc network is introduced.The fractal method is used to analyze and study the network node clock noise.Kalman filtering is a method often used to describe the clock difference of network nodes.The Kalman filtering method of the clock model can describe the noise state of the node clock.Analyze the non-normality of the clock noise sequence.The fractal characteristics of jet lag noise are introduced.The fractal characteristics can be used to estimate the covariance matrix of Kalman filtering,thereby improving the efficiency of the filtering algorithm.(3)Analyze the two-way time synchronization error of RBS,TPSN,Tiny / MiniSync three time synchronization algorithms.Using the fractal prediction method of network node clock error,the noise prediction model is established by using the fractal characteristics of the clock error between nodes,and the synchronization accuracy is improved by improving the estimation accuracy of the noise covariance matrix.Through simulation comparison,the prediction accuracy of the relative clock difference between nodes is higher than that of the first-order polynomial model and the gray model under the fractal model.Three network nodes were used to verify the performance of RBS algorithm,TPSN algorithm and Tiny / Mini-Sync algorithm.According to the network time synchronization model based on clock difference prediction,by describing the fractal characteristics of clock difference noise,the influence of noise on the time synchronization accuracy of network nodes is reduced.In terms of convergence speed,the network synchronization algorithm based on time difference prediction is faster than the three time synchronization algorithms RBS,TPSN,Tiny / Mini-Sync,and there is no loss of accuracy.