An Application of Wireless Sensor Networks in Underground Coal Mine

Abstract

Wireless sensor networks (WSN) are good at security monitoring in coal mine. It is able to rapidly detect diverse parameters, which can reduce human and material losses. And it is an important application that has commercial potential. Coal mine pivotal parameters include Dust Density (Dust), Temperature (Temp), Wind Speed (WindS), Gas Density (GasD) and Carbonic Oxide Density (COD). The data collected by the sensors are sent to the sink node to be processed with information fusion technology. This work presents a strategy for the classification of coal mine status based on sensed data by WSN and the use of unsupervised neural network-the Self-Organizing Map (SOM). The SOM application classifies the coal mine environment into four clusters. An experiment confirms the effectiveness of the proposed approach.