The invention belongs to the field of communication, and particularly discloses a tactical sign language recognition glove system based on deep learning and a sensor technology and an implementation method. Comprising a Raspberry Pi module, an arduino development board, a V5 expansion board, a power supply module, a switch module, a PC end interface module, a receiver earphone end, a bending sensing module, a gyroscope sensing module, a pressure sensing module and a satellite positioning module. According to the invention, a single-label multi-classification neural network gesture recognition model is established based on Keras, and a glove collection data-based sign language recognition system is established based on a sensor technology, so that sign language transmission information can be utilized in real time, and information exchange based on sign language recognition can be realized; remote information transmission between users can be realized, and accurate information interaction of the users under the blocking of obstacles can be established; the position condition of each user can be obtained in real time; automatic selection of emergency measure plans of emergency situations can be established, meanwhile, emergency communication of users is facilitated, and then the complete, accurate and high-real-time communication function of the whole system is achieved.