Research on test system of open-pit mine microwave network based on convolutional neural network

Due to the complex environment of open-pit mines, in order to monitor illegal mining and other activities in real time, it ensures the timely synchronization of massive data to the microwave network of open-pit mines, and makes its functions meet expected needs. Therefore, this paper proposes a microwave network testing system for open-pit mines based on convolutional neural network. It designs a main chip hardware structure based on the top-down principle and a communication module for the shielding chamber to achieve lossless acquisition of microwave network data by the system. In the software design of the system, convolutional neural network is introduced to perform feature analysis on the microwave network data obtained from the shielded chamber in open-pit mines. Combined with multi-dimensional clustering processing of feature vectors, erroneous features in the network data are captured to obtain more accurate test results, completing the overall design of the testing system. According to the example testing and result analysis, the network testing function of the system meets the expected requirements, the network testing results are relatively accurate, and the synchronization time is stable within the range of 16-18 ns, with excellent time synchronization accuracy performance and good practical application prospects.