The Research And Implementation Of Forest Fire Monitoring And Warning System Based On Internet Of Things

Forest fires are a very harmful natural disaster.it often brings about the consumption of large forest resources and the far-reaching impact on the surrounding agricultural animal husbandry.At the same time,Forest fires is usually difficult to deal with and rescue due to the special situation of the scene.Therefore,the monitoring and warning of forest fire indicator environment has become a highly valued problem between many countries.With the rapid development of technology,the emerging technologies of the Internet are also increasing.Since most of the forests are located in remote areas,the cost of deployment and maintenance of traditional networks is high,and the problems of inconvenient supervision and insufficient power supply have led to the lower advantages of traditional networks in forest fire monitoring.The emerging wireless sensor network technology combined with the Internet of Things architecture has its unique advantages in special areas such as unattended remote environment monitoring and disaster suppression.The Zigbee technology is a standard wireless network protocol designed for application scenarios with low data consumption requirements,low power consumption,low cost,low latency,flexible networking,and security features.Its emergence combined with the various needs of the forest scene makes this technology have an incomparable advantage in forest fire monitoring.At the same time,due to the variable and complex geographical factors of forest fires,therefore,the accuracy of determining whether a fire occurs by a single sensor is often low.It is easy to cause waste of rescue resources when applied to actual scenes,and multi-data indicators must involve multi-factor data fusion.Therefore,this problem is especially important in the case of forest fires.This paper focuses on improving the direction of warning accuracy of forest fire monitoring and alarm system.Since the indicator information in this scenario is typical “uncertainty” information,based on the research of multi-sensor data fusion technology,the DS evidence theory algorithm is researched and analyzed,and the basic probability assignment problem based on the existing DS evidence theory and Conflict data leads to related improvements in decision-making failures.Firstly,propose a first-level fusion algorithm based on fuzzy inference to solve the basic probability assignment problem in DS evidence theory algorithm.Extend the traditional single eigenfunction assignment method to double feature fuzzy reasoning method,improve the correlation between basic probability and feature.And finally use Matlab to simulate,the simulation results were compared with the traditional methods.Then,propose the second-level fusion algorithm based on weighted optimization to solve the problem that the conflict data in the DS evidence theory algorithm is easy to cause the decision-making invalid problem.Use the fuzzy conflict-progressing concept to decentralize the global conflict into local conflicts and the local conflict as the weighted index to optimize the basic probability.Reduce the impact of conflict data on the overall integration,and also use Matlab to simulate and analysis on the improved algorithm.On this basis,this paper conducts a detailed demand analysis and program construction for the forest fire monitoring and warning system,implements the system from the perspective of the three-layer architecture of the Internet of Things,and finally verifies the basic functions and improvements of the system.The applicability and feasibility of the algorithm.