Abstract The rapid development of computer network technology has brought great convenience to all walks of life and made the transmission of various information faster. A secure and reliable computer network environment has a critical impact on the functioning of the computer network. In this paper, we derive and analyze the detection probability, false alarm probability and judgment threshold of the new algorithm based on the results of random matrix theory based on the feature-fitting superiority algorithm, which can exhibit better detection performance. In the framework of matrix theory, the superiority algorithm reduces the problem of network security to some extent and improves the performance relative to other fitted superiority detection algorithms. For common threats to network security, give the corresponding ideas to deal with network security issues from a technical perspective. The detection rate and false alarm rate are calculated by testing the total number, the number of intrusion bars, the number of detected intrusion bars, and the number of false alarm bars detected by the system, respectively. The test results show that the average detection rate is still greater than 94%, the average false alarm rate has improved, and the average false alarm rate is controlled at about 3.1%, which can fully meet the system requirements. Give reference to further enhance the reliability and stability of network security systems.