Network segmentation has long been employed as a perimeter-based security model designed to block external threats. However, as cloud computing, remote work, and mobile work environments continue to expand, the limitations of the existing structure, w...
Network segmentation has long been employed as a perimeter-based security model designed to block external threats. However, as cloud computing, remote work, and mobile work environments continue to expand, the limitations of the existing structure, which depends on implicit trust within internal networks, have become increasingly evident. Recent policy improvements related to network segmentation further indicate the need for a transition toward a more flexible and risk-based security framework, a direction that is closely aligned with the Zero Trust principle of continuous verification. Despite this shift, studies on the technical applicability and structural interpretation of Zero Trust within network segmentation environments remain limited.
This study analyzes the structure of network segmentation, along with its authentication and access control mechanisms, from an architectural perspective. In addition, it examines the structural differences between the traditional network segmentation model and Zero Trust principles.
Based on this analysis, the study identifies inconsistencies between the two control structures and evaluates the potential for technical compatibility between these security paradigms in the context of ongoing regulatory change.