Abstract: In order to solve the problems of low accuracy，slow response speed and limitations of sin⁃ gle sensor monitoring in gas concentration monitoring in coal mine，this paper proposes an online monitor⁃ ing method of gas concentration in coal mine based on multi-source information fusion. In this study， multi-sensor information data fusion technology is used to integrate multi-source data collected by meth⁃ ane sensor model IYFAS-A8FA8，temperature sensor model IYHFT-A7FG8 and air volume sensor model IGHFA-A7F8 A. Firstly，the abnormal data is eliminated by moving average line processing method，and the noise is filtered by orthogonal wavelet transform technology. Then，the double fusion mechanism is used to realize the accurate monitoring of gas concentration，and the corresponding real-time monitoring and early warning system is constructed. Taking a coal mine as the research object，the experimental verifi⁃ cation is carried out. The results show that the monitoring sensitivity of the method is above 95%，up to 99.65%，which is 31.46% higher than that of the monitoring method based on BP neural network，and 22.45% higher than that of the monitoring method based on deep learning. The monitoring value is highly consistent with the actual gas volume fraction and is within the confidence interval. The monitoring method proposed in this paper not only effectively improves the accuracy and reliability of gas concentration moni⁃ toring，but also solves the shortcomings of traditional monitoring methods. It can also avoid safety risks in a timely manner through graded early warning，providing a strong technical guarantee for coal mine produc⁃tion safety. At the same time，it provides a practical reference for the application of multi-sensor informa⁃ tion fusion technology in the field of coal mine safety monitoring，and has a good application prospect.

Key words: multi-source information fusion, gas concentration, on-line monitoring, temperature sen? sor, methane sensor, air volume sensor