Design And Implementation Of Dust Intelligent Early Warning System Based On LoRa

In recent years,with the rapid development of the social economy,the problem of dust pollution has become more and more serious.Dust pollution not only restricts economic production,but also endangers people's lives.At present,the dust monitoring system has problems such as short transmission distance,poor anti-interference ability and insufficient intelligence level.Aiming at the above problems,combined with sensor network technology,long-distance(LoRa,Long Range)wireless transmission technology and embedded development technology,a dust intelligent early warning system based on LoRa is designed.The system is mainly divided into three parts,namely data acquisition subsystem,data transmission subsystem and intelligent host,which realizes the function of dust site monitoring.The research contents of each part of the thesis are as follows:Firstly,in the data acquisition subsystem,the data terminal obtains the temperature,humidity and dust concentration data of the dust scene through three kinds of sensors,and guarantees the safety of the dust scene through the alarm device and the control device.In this part,the system completes the software and hardware design of the temperature measurement cable,humidity detection module and dust concentration detection module.In order to ensure the accuracy of the temperature data,the system realizes multi-point temperature measurement through the DS18B20 temperature measuring cable.Aiming at the low efficiency of serial number traditional binary tree search algorithm in single bus DS18B20 temperature measurement cable,an optimized binary tree search algorithm is designed based on the identification of the same family code and number of bits in different DS18B20 serial numbers of single temperature measurement cable.The algorithm can ensure search accuracy and improve search efficiency.Secondly,in the data transmission subsystem,the LoRa terminal implements data transmission between the data terminal and the intelligent host.In this part,the system completes the LoRa terminal hardware and software design and low power design.By analyzing the advantages and disadvantages of various network structures and combining the advantages of LoRa wireless transmission technology,a low-power long-distance data transmission scheme based on star network structure is designed.The main contents of the scheme include:the sleep mechanism based on channel activity detection,the low-power wake-up mode combined with the two methods,the orderly uploading data according to the module number and the multi-process efficient operation using the Contiki embedded system.Through the established low-power calculation model,under normal circumstances,the battery-powered terminal of the system can work for at least one year.Finally,in the smart host,the smart host implements the reception,storage and prediction of sensor data.In this part,the system completes the construction of a platform based on embedded ARM+Linux and designs a prediction model based on local weighted linear regression algorithm.Combining the historical correlation with the sensor data at the next moment,a prediction model based on local weighted linear regression algorithm is designed by assigning different weights to the data near the predicted points.The comparison between the model and the prediction model based on linear regression algorithm verifies that the model has better prediction effect.Through the test and analysis of LoRa terminal and system functions,the results of multiple sets of tests show that the dust on-site monitoring system designed by this system has a long transmission distance and low system power consumption,and has the function of intelligent early warning.The system solves the problems existing in the field of dust detection and has strong application and promotion value.