Research and Application of Situational Awareness in Hospital Network Security

With the rapid development of medical informatization, the diversification and complexity of hospital network services have increased significantly, raising higher demands for network security. Current hospital network security defenses primarily rely on traditional rule-based interception and single-layer protection, lacking comprehensive and proactive situational awareness capabilities. This paper aims to develop a comprehensive situational awareness model for hospital network security to enhance threat identification, early warning, and prediction. First, to address the complexity of hospital networks and facilitate element acquisition, principal component analysis (PCA) is employed to reduce the dimensionality of network access traffic data, removing redundant attributes and improving the speed and accuracy of threat classification and prediction. Second, a multi-dimensional hierarchical situational awareness index system based on multi-source data is constructed, with indexes quantified using entropy weighting and hierarchical analysis, enabling an accurate evaluation of hospital cybersecurity. Finally, a time series-based prediction method is proposed, utilizing the seasonal autoregressive moving average (SARMA) model to effectively forecast the hospital's network security situation, providing practical guidance for real-world security management.