Research On Network Time Synchronization Using Beamforminng Antennas

With the development of wireless communication technology,wireless ad hoc network is widely applied in real scenes.As a kind of peer to peer network,it can work without central infrastructures.Compared with the Omni-directional antennas,the beamforming antennas,as a kind of smart antennas,have the advantages of long transmission distance,strong anti-interference ability and high spatial reuse coefficient,so that the combination of beamforming antennas and wireless distributed networks will bring the performance improvement.However,due to the introduction of beamforming antennas,it also brings about many problems,which is difficult to complete clock synchronization among the nodes equipped with beamforming antennas,resulting in decreased performance of network protocols based on time synchronization.While the mobile nodes will also cause dynamic changes of the network topology,the beamforming antennas need to take more time to adapt it.Time synchronization is the supporting technology of wireless distributed network.Many MAC layer protocols and the applications are based on the network clock synchronization.In this paper,we proposed an asynchronous scanning synchronization algorithm for quasi-static network using beamforming antennas.The algorithm can be divided into two phases: root election and synchronization.In the election phase,it selects the node owning the maximum neighbours as the root.In the synchronization phase,the nodes synchronize with the root by pairwise communications and promote the whole network synchronization.We analyze the cost of the election strategy and the error of the synchronization phase.We also raise a level optimization strategy using spanning tree,which can reduce the overhead significantly.Simulation results show that the proposed algorithm can reduce overhead and address time synchronization problems of beamforming network efficiently.For the dynamic network topology,we put forward a clustering synchronization with topology maintenance,which is divided into four phases: election,clustering,synchronization,maintenance.Election is to elect the node with the maximum neighbors as the root and then divide the network into clusters in asynchronous and directional network,meanwhile realize network synchronization.Synchronization phase stabilize the clock synchronization and the last phase is maintaining topological relationships.The simulation,by The Opportunistic Network Environment simulator,shows that the algorithm has excellent synchronization performance,and can adapt to mobile ad hoc networks well.