Event-driven Energy-efficient Distributed Filtering And Fusion

Wireless sensor networks(WSNs)is currently a top researeh field with a wide range of applications.It can sense the physical world,such as sensing the temperature and humidity changes of environment,traffic control etc.The development of WSNs technology makes human's life more intelligent and achieve the communications between human world and physical world.However,the WSNs have three main weakness:the nodes equipped with limited energy,large-scale network and difficult to replace or recharge the node's battery.So,how to prolong the lifetime of network maximaly is always the research hotpot.1)A novel distributed Kalman-consensus filtering algorithm based on event-driven mode is discussed.In order to adjust the accuracy adaptively and reduce the amount of data transmission among nodes when the algorithm adopts the event-driven ideas.Then,the convergence of the algorithm is analyzed based on graph theory and matrix theory.2)Updated consensus distributed Kalman filtering algorithm based on clustering model is studied.By introducing clustering model to reduce the amount of data transmission and shorten the communication distance among nodes.Moreover,the update items of algorithm add a consensus items to further improve the estimation accuracy.Then,the convergence of this new algorithm is proved by applying to Lyapunov method and matrix theory.3)A distributed cluster consensus filtering algorithm based on event-driven is also discussed.Combined with the event-driven ideas and clustering model,that makes the algorithm also can guarantee the accuracy and further conserve energy.Then,the convergence of this new algorithm is proved by applying to Lyapunov method and matrix theory.