Research On Key Technologies Of LoRa Application

Low Power Wide Area Networks(LPWAN),designed for low bandwidth,low power consumption,long distance and large number of connected Internet of Things applications,has emerged to meet the needs of more and more remote Internet of Things devices.LoRa has attracted wide attention in many LPWAN technologies due to its low cost,easy networking and high performance.In recent years,LoRa technology has been widely used in the fields of smart city,smart industry and smart home.This paper mainly studies the deployment of LoRa gateway and indoor location based on LoRa.In the LoRa gateway deployment problem,how to reduce the power consumption of the terminal in LoRa network by selecting the location of the gateway under the condition of meeting the business needs is the optimization goal of the problem.Specifically,based on the application background of LoRa gateway deployment of LoRa intelligent water meter network,a method of LoRa network gateway deployment based on immune algorithm is proposed.Firstly,the optimization model of LoRa intelligent water meter network gateway deployment is established,and the power optimization problem is transformed into the distance optimization problem,including distance constraints and capacity constraints.Then,the inmune algorithm is used to solve the problem,which is compared with the traditional genetic algorithm.The experimental results show that the proposed LoRa network gateway deployment method based on immune algorithm can converge quickly and obtain the sub-optimal gateway deployment scheme.It provides a fast optimization method for the location deployment of LoRa gateway and effectively solves the problem of gateway deployment location selection in the construction of LoRa intelligent water meter network.In the research of indoor positioning based on LoRa,we design and implement the indoor positioning protocol and prototype system based on LoRa?and then propose an indoor positioning algorithm based on LoRa fingerprint and Support Vector Regression(SVR).Firstly,based on LoRa SX1280 module,the design and implementation of indoor positioning protocol flow and data frame format are completed.Then,aiming at the problem that the traditional RSSI fingerprint based on wireless signal(received signal strength indication)and SVR indoor positioning algorithm have low positioning accuracy,two improvements are proposed:1)LoRa ranging fingerprint is added to improve fingerprint stability;2)In the process of establishing fingerprint database,Gauss filtering is used,while median filtering is used in online positioning.Both of them aim at preprocessing fingerprints and eliminating gross errors of fingerprints.Finally,field experiments in real indoor environment are carried out by using the LoRa indoor positioning system and fingerprint positioning algorithm.The experimental results show that the cumulative probability of positioning error within 1m is 78.5%and that of positioning error within 3m is 90%.In the first aspect of improvement,the positioning accuracy is improved by 40%after adding LoRa ranging fingerprint.In the second aspect of improvement,the positioning accuracy of pre-processing with Gauss filter and median filter is increased by 38%compared with the traditional support vector regression algorithm.After these two improvements,the positioning accuracy is improved by 63%,which proves that the two improvements of the algorithm are effective.