Research On Low-energy Poisonous Gas Monitoring And Data Analysis In Wireless Sensor Networks

Since the rapid development of wireless sensor networks,sensor monitoring technology can be seen everywhere,for example,monitoring poisonous gas,it can use the concentration detection function of gas sensors.Once a leak occurs,the concentration data detected by the sensor will be higher than normal value.The sensor node’s power is usually supplied by the battery,but the battery that can’t be replaceable is its fatal shortcoming.When a node’s battery energy is exhausted,the entire network’s life cycle will be terminated.Therefore,how to reduce energy consumption and extend the network life is the focus and difficulty of the monitoring task.Moreover,how to deal with the abnormal information detected after a leak has occurred and guide the user to make reasonable decisions is also an integral part of the monitoring task.Therefore,the research content of this article can be mainly divided into the following two parts:(1)The low-energy mechanism of poisonous gas monitoring: the idea of having the sensor nodes constantly detect is obviously contrary to the idea of saving energy,so people spontaneously think of cycle detection.There are two main approaches in previous work,one is to have all nodes periodically detected,and the other is to make a part of representative nodes periodically detect,the former will bring too much energy consumption,while the latter makes the detection results inaccurate.Taking comprehensively into account two factors of energy saving and accuracy,this paper adopts a new low-energy mechanism—activation mechanism.The mechanism is based on the rational selection of the head node according to the periodicity of the remaining energy,let the head node periodically detect,and other nodes keep sleep.Then comparing the concentration value detected by the head node with the concentration threshold.After the comparison,once it exceeds,other nodes within its communication range are activated by sending data packets,and the activated neighbor nodes continue to detect.At the same time,the abnormal node reports information to the sink node through the shortest path.(2)The data analysis of abnormal nodes: since the abnormal data collected on the sink node is very useful,so this paper analyzes it from three aspects,boundary node extraction,isoline generation and leakage source location estimation,respectively.To extract boundary nodes,the position coordinates of all abnormal nodes are divided into different intervals in ascending or descending order in a certain direction,then in each interval,the nodes with the largest and smallest coordinate values in the other direction are found,these nodes are regarded as boundary nodes;As to the second point,three existing interpolation algorithms are used to interpolate,namely cubic,nearest and invdist;Besides,reusing the position coordinates of the maximum concentration values obtained by the three interpolation algorithms to estimate the location of the leakage source,and a better one is selected after comparison.This estimation method is also proposed for the first time.By simulating the situation within a short period of time after the leak occurred,the theoretical analysis and simulation results demonstrate the energy efficiency and technical feasibility of the proposed method.