Research And Application Of Debris Flow Prediction Based On Sequence Similarity Location Algorithm In WSN

Debris flow disasters usually occur in mountainous areas with steep terrain.The destructive power of disasters has caused serious threats to the production and life of mountain residents.As an important disaster reduction method,debris flow prediction has become the focus of research at home and abroad.Traditional monitoring methods cannot achieve good prediction results,and wireless sensor networks obtain node locations through positioning technology.They can monitor rainfall by nodes in real time and determine the location of disasters when the environment is abnormal.Aiming at the problems in rainfall-type debris flow prediction,this paper designs and realizes a debris flow prediction system based on WSN technology,which achieves the purpose of all-weather and real-time monitoring and provides a strong guarantee for the life and property safety of people in mountainous areas.The locations of nodes are the basis for obtaining other important information and are essential for the good performance of the entire network system.Positioning technology has become one of the key contents of WSN research.The main research contents of this paper are as follows:First of all,this article analyzes the significance of using the positioning technology in WSN to construct a debris flow prediction system,briefly introducing the basic knowledge of WSN,and focuses on the relevant positioning technology.Secondly,in view of the problem that signal propagation in three-dimensional space is easily affected by obstacles,resulting in low positioning accuracy.A three-dimensional positioning algorithm based on sequence similarity(3D-LASS)is proposed.The algorithm used the properties of the Voronoi diagram to divide the space into multiple continuous and dense polyhedral sets.The algorithm used the weighted median filter method to modify the signal value and combined the coplanarity between nodes to further optimize the order sequence of unknown nodes.The algorithm introduced virtual node and determined the order sequence of virtual nodes and beacon nodes and constructed the nearest neighbor sequence list of unknown nodes.The algorithm used the similarity between each type of sequence to estimate the position of the node with weight.The algorithm proposed in this paper corrected the signal while considering the spatial topology of the node,overcoming the influence of errors caused by model conversion,and improved the positioning accuracy while reducing the amount of calculation.Finally,the positioning algorithm proposed in this paper has been applied to the debris flow prediction.The debris flow prediction system has been designed and realized.According to the needs of the system,the CC2530 chips equipped with rain sensor are used on the terminal node to complete the rainfall data collection.The client system adopts the B/S architecture and uses the Java language to complete the development of the system.The host computer realizes the functions of node positioning,real-time reception of environmental data and dynamic display.At the same time,system completes the functions of data analysis and safety warning according to the established debris flow prediction model.Tests have verified that the system is accurately positioned and also can operate efficiently and stably,meeting the actual demand for predicting debris flow.