Font Size: a A A

Research On Time Synchronization Algorithm For Wireless Sensor Networks

Posted on:2017-07-29Degree:MasterType:Thesis
Country:ChinaCandidate:S LiFull Text:PDF
GTID:2348330482486851Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
With the in-depth study of wireless sensor network(WSNs),WSNs technology has been developed rapidly.As one of the supporting technologies in wireless sensor networks,time synchronization technology is the foundation of the technology of node localization,data fusion,and node periodic wake up.On the basis of summarizing the characteristics of wireless sensor network,key technology and the current classical time synchronization algorithm,a hybrid time synchronization algorithm based on dynamic equilibrium hierarchical tree is designed.Then the lightweight time synchronization optimization algorithm based on self-learning compensation frequency drift is designed.In this paper,the time synchronization algorithm in WSNs is summarized and compared,and we found that the performance of Timing-Sync Protocol for Sensor Networks(TPSN)and Flooding Time Synchronization Protocol(FTSP)algorithm in their application scenarios is better than others,respectively.However,there is a potential risk when many child nodes run out caused by father node died,so The whole network need to be initialized again.Also,both TPSN and FTSP cannot synchronize on demand.In this paper,the dynamic topology of FTSP algorithm is applied to the hierarchical tree topology,a hybrid time synchronization protocol based on the TPSN algorithm is designed.Simulations present that the proposed algorithm reduces the channel congestion and the imbalance of different path synchronization errors in FTSP algorithm,compared with the TPSN algorithm,it can improve the synchronization accuracy and robustness,and it can be synchronized on demand.TPSN time synchronization algorithm is a widely used global time synchronization algorithm,its security and stability are better than the one-way synchronization algorithm.But such algorithms cannot compensate for clock frequency drift,it will cause low period of sync and high energy consumption under high accuracy.We propose a self-learning based on TPSN optimization algorithm,taking the TPSN algorithm time clock error as a function of time and the frequency drift,The clock frequency drift is obtained by means of geometric deduction,thus avoiding the above problems.The simulation results show that the proposed algorithm can compensate the error caused by the clock frequency offset,and thus significantly prolong the synchronization interval.
Keywords/Search Tags:Wireless sensor networks time synchronization, dynamic equilibrium hierarchical tree, robustness, compensate clock frequency
PDF Full Text Request
Related items