Research On Distributed Clock Parameters Modeling In Wireless Sensor Networks

With the rapid development of wireless sensor networks(WSNs), its applications become more widely. Most of these applications require collaborative execution of a distributed task amongst a large set of synchronized sensor nodes, Therefore, clock synchronization has become the basis of many applications of WSNs. In recent years, distributed clock synchronization has attracted greatly concerned and has been a hot research topic. Distributed clock synchronization algorithms can not only overcome the lack of less scalability of traditional clock synchronization protocols, but also run a fully-distributed operation, which makes the algorithm more practical. Unfortunately, there are three problems in the existing research works in terms with distributed clock synchronization algorithms, including the low synchronization accuracy, slow convergence speed and high energy consumption. To make up the lack of these algorithms, from the perspective of control theory, this paper describe the dynamic process of clock synchronization under the framework of distributed clock synchronization state model, propose corresponding protocols and provide system theory and algorithm simulation analysis. The main contents of this paper are summarized as follows:Firstly, a distributed clock synchronization state space model is designed. In the first place, based on the integral clock model, clock parameter status update model is established; then, based on the clock synchronization mechanism of two-way message exchange, the clock relationship between adjacent nodes over the whole network is established, according to the clock relationship, local timestamp measurement model is established; lastly, taking status update model and local timestamp measurement model as the core, the distributed clock synchronization state space model is constructed. The model is scalable with network size.Secondly, in the presence of unreliable factors in WSNs, in order to improve the anti-interference ability of the system, make up the estimation error and improve the accuracy of clock synchronization, from the perspective of control theory, this paper take distributed clock synchronization state model as a framework, distributed clock synchronization state space model is established based on networked control perspective.Thirdly, based on the networked control perspective–based distributed clock synchronization state space model, networked control perspective–based distributed clock parameters tracking algorithm is proposed. This algorithm mainly relates to the design of estimate and control. This paper develops a distributed Kalman filter algorithm for clock parameters estimating and tracking about estimation algorithm. This paper develops a distributed single-step optimal control algorithm about control algorithm, which is scalable with network size. The proposed tracking algorithm only requires each node to exchange limited information with its direct neighbors, thus is energy efficient.Fourthly, the simulations show networked control perspective–based distributed clock parameter tracking algorithm has higher synchronization accuracy, faster convergence speed and better scalability.