Research On Energy Balanced Routing And Node Location In ZigBee Network

The rapid development of the Internet of Things and wireless sensor network technology has greatly changed our way of life,enriched the application scene and improved the application range.ZigBee technology is widely used in wireless sensor networks for its advantages of low cost,low power consumption and flexible networking.In this paper,ZigBee technology is adopted for networking to build a forest fire monitoring system,so as to monitor the occurrence of fire in real time and locate the fire area,so as to take effective measures to control forest fires.The wireless sensor network composed of ZigBee uses battery power and its own energy is limited;Using the received signal sensitivity of ZigBee module for ranging and positioning also has a large error.Therefore,the following contents are studied in this paper:(1)In order to prolong the working cycle of wireless sensor network,a LEACH-COMR(Clustering Optimization Multihop Rooting)algorithm based on LEACH protocol is proposed in this thesis.During network initialization,the number of optimal cluster heads is calculated,and the residual energy of nodes and the location information of nodes are fully considered in the process of cluster head selection.In the clustering stage,by limiting the number of members in the cluster,the cluster heads with the most remaining energy are selected to be added to the redundant nodes to prevent the appearance of the maximum and minimum clusters.In the data transmission stage,relay parameters are introduced and appropriate relay nodes are selected to forward cluster head fusion data to the base station.The simulation results show that the LEACH-COMR algorithm can effectively reduce the excessive consumption of node energy and extend the operating cycle of wireless sensor networks.(2)To solve the problem of RSSI(Received Signal Strength Indicator)distance measurement error,this paper proposes a RSSI ranging method based on BP neural network.Two frequency bands of 433 MHz and 2.4 GHz were used to obtain the corresponding RSSI values,and the obtained values were filtered by Gaussian filtering to remove the outliers,and then substituted into the logarithmic shadow path model for distance calculation.Average the distance values of the two frequency bands measured.Propagation(Back Propagation)neural network was used to iteratively optimize the processed range.The results show that the predicted distance value obtained by BP neural network can fit the real value well,and the root mean square error of ranging is 1.09 m,which greatly improves the accuracy of RSSI ranging.(3)According to the research background of the paper,build the architecture of forest fire monitoring and design function modules.The wireless sensor network is composed of terminal,router and coordinator.The terminal is equipped with temperature and humidity sensor and smoke sensor for data acquisition work;In addition to collecting information,the router also forwards the data sent by other terminals.The coordinator is responsible for collecting,summarizing,and uploading data information to the server.In addition,in order to improve the communication distance of ZigBee nodes,the RF module layout is redesigned and a power amplifier is added to increase its transmitting power.The network function is developed on the basis of improving the communication distance,realize the control of each terminal module.