Energy Efficient Distributed Kalman Consistency Filter Design

Wireless sensor network(WSN)is a multi-hop wireless network which is self-organized by a large number of wireless sensors scattered in the work area by wireless communication.Wireless sensor network is an information sensing and data acquisition system,which can realize the perception and acquisition of physical information in the physical world.It is an important research filed of information science and technology,and plays an important role in sensing,monitoring,remote control and intelligent applications.A group of sensors with limited functions to complete a large-scale sensing and monitoring task is an important feature of wireless sensor network.The limited capacity mainly refers to sensor nodes with the limited computing resources,storage,communication,energy and so on.Therefore,developing suitable methods to overcome these shortcomings becomes one of research hot spot in wireless sensor network.This dissertation studies the energy saving problem of wireless sensor network after the designed distributed filtering algorithm ensures enough estimation accuracy.From the perspective of control network data transmission and control network topology,several distributed Kalman consensus filters are proposed.The main work is summarized as follows:1.A resource-aware distributed Kalman filter is designed with stochastic communication based on site-percolation model.The wireless sensor network is modeled as a two-layer topology,combining clustering algorithm and site-percolation model to short the communication distance and reduce the communication complexity among nodes,and then a novel distributed Kalman filter algorithm is proposed.The convergence of the algorithm is analyzed by the matrix theory.Next,the dichotomy method is used to find the optimal active probability to reduce the unnecessary energy consumption under the given filtering accuracy.Finally,the effectiveness of the algorithm is verified by simulation example.2.Event-driven distributed Kalman-consensus filter with time-efficiency window is designed.The concept of the time-efficiency window is proposed based on the relationship between time and information efficiency.And then,an improved event-driven mechanism combing with the time-efficiency window is also designed.Consequently,a novel event-driven distributed Kalman consensus filter with time-efficiency window(TEDKCF)is presented.By adjusting the length of the time-efficiency window,the historical data stored on each sensor are used fully.And then,the life of WSN is prolonged on the condition that the similar filtering accuracy is achieved than that of the event-driven distributed Kalman-consensus filter(EDKCF).Furthermore,the sufficient conditions of mean square stability of the designed filter are obtained by using the second Lyapunov method.Finally,simulation results show the effectiveness of the proposed filter.3.A new type of distributed event-driven Kalman consensus filter with double communication radius is designed for energy saving.The communication radius of nodes in wireless sensor network is designed to self-adjust.Based on the event-driven condition of estimation accuracy,the communication topology decision-making mechanism is designed to form the second level event-driven condition for the energy saving.The decision-making mechanism dynamically selects the communication radius,so as to reduce the number of fusion nodes effectively and then extend the lifetime of sensor nodes.Finally,the proposed energy-saving algorithm is shown by a simulation example.