Abstract: Underground coal seam mining causes the deformation, failure, and collapse of the overlying rock mass of the coal seam. Suppose the monitoring and early warning of overburden deformation and failure during mining or the emergency response are inappropriate. Mining disasters such as roof falls, water inrush, gas outbursts, and rock bursts will often occur in the working face. To ensure the intrinsic safety of the mining face, it is of great significance to implement full real-time dynamic sensing and whole process intelligent warning of the overlying strata deformation caused by coal mining. A large number of field monitoring test results show that the deformation and failure of the overlying rock and soil mass of the coal seam are discontinuous and nonlinear. Therefore, how to break through the bottleneck of spatio-temporal discontinuous observation of overlying rock deformation and obtain real-time and accurate multi-field data of overlying rock and soil mass of the coal seam has become a critical link in the mechanism research for mining overlying rock deformation and failure. Based on the distributed fiber optic sensing technology, this paper systematically studies the deformation field, temperature field, seepage field, and geoelectric field of the overlying rock and soil mass during the coal mining process, introduces the mechanism, observation technology, research status, and development trend of each physical field in detail, and analyzes the research direction of mine engineering microseismic based on DAS technology, to provide theoretical support for the realization of “transparent” exploration of mine underground space. A safety guarantee system for coal seam overburden during mining integrating sensing, transmission, processing, early warning, and decision-making emergency is established through constructing a fiber optic sensing neural network combined with surface and downhole. The system consists of five modules: real-time sensing, reliable transmission, intelligent processing, instant early warning, and emergency decision-making, which can realize the intelligent sensing of the whole process of overburden deformation and failure during mining as well as the correlation analysis and evaluation of multi-field real-time data. The research results can also provide a reference for the monitoring and evaluation of the geological environment within the mining area and the restoration and treatment.