Deep Learning-Driven Network Security Situation Awareness Method in 6G Environment

Abstract

With the rapid advancement of computer technology, the incidence of network security breaches has significantly increased, leading to a corresponding rise in cyber-attacks. Traditional security defense mechanisms exhibit inherent limitations, characterized by their reactive nature and limited efficacy against unknown threats. In addition, these mechanisms often lack coordination among different components, further diminishing their overall effectiveness. In response to these challenges, this paper proposes a deep learning-driven network security situational assessment (DL-driven NSSA) method specifically designed for the 6G environment. Initially, a deep autoencoder (DAE) model is constructed to identify various types of attacks within the network. Subsequently, the model undergoes rigorous testing to calculate attack probabilities, determine impact scores for each attack, and compute the overall network security situational value. Experimental results demonstrate that the proposed DAE model outperforms existing models in terms of accuracy and recall rate, thereby enhancing the precision and reliability of assessment outcomes.