Research On Consensus Methods For Apron Crowd Sensing Network

Smart apron is an important part of smart airport construction,and the intelligent perception of apron is an important direction.A network composed of most wireless sensors and sensing nodes integrated in smart terminals is called the Apron Crowd Sensing Network(APCSN),which realizes the perception and collection of apron production information,and promotes apron run efficiently.Time synchronization is the technical support for practical applications such as data fusion,coordinated monitoring,and node positioning by apron crowd sensing network.Consensus time synchronization algorithm has been widely studied in recent years due to its superior performance such as robustness and scalability.However,current consensus-based algorithms are time-consuming,and there is a problem of slow convergence speed due to iterative operations.Therefore,it is very meaningful to study a consensus time synchronization algorithm with faster convergence;at the same time,the topology is an important factor which affect the convergence speed.Therefore,from the perspective of network topology,improving connectivity and aiming at complex topologies,the paper proposes a corresponding improvement method for consensus time synchronization.Aiming at the problem of the slow convergence speed of the consensus time synchronization algorithm,from the perspective of improving network connectivity,a Fast Convergence Average Time Synch(FCATS)algorithm based on virtual links is proposed.Based on graph theory and consensus theory,the convergence speed can be improved by increasing the algebraic connectivity of the network,and the algebraic connectivity can be improved by increasing the network connectivity.Based on the above analysis,virtual links are added between double-hop neighbors,and double-hop neighbor nodes of nodes are added,and network connectivity is improved.Then,considering the impact of communication delays,the relative clock skew,logical clock skew,and clock offset estimation models are established.Based on the clock information of single-hop and double-hop neighbor nodes,the clock parameters of the nodes are updated,and the algorithm uses consensus iterations to achieve clock synchronization.The simulation experiment verifies the feasibility of the algorithm,and the convergence speed of the fast convergence consensus time synchronization algorithm based on the virtual link is improved compared with other algorithms.Community is an important feature of the apron crowd sensing network with the increasingly complex topological characteristics,its characteristics is that the links within the community are close,and the links between the communities are sparse.The sparse link reduces the convergence of the consensus algorithm,and makes the convergence block which is that the synchronization speed within the community is faster,but the synchronization speed between the communities is inconsistent,so that the overall network convergence speed decreases.To solve this problem,a Community Division based Consensus Time Synch(CDCTS)algorithm is proposed.Considering the influence of communication delay,model of the estimation of relative clock skew,logical clock skew,and clock offset is established.Through the synchronous coordination of inter-group consensus synchronization and intra-group consensus synchronization,the convergence speed of consensus is accelerated.The simulation experiment verified the feasibility of the algorithm,and convergence speed and energy consumption of communication of the consensus time synchronization algorithm based on community division is improved compared with other algorithms.