An adaptive NLP-driven access control framework for Northbound Interface in Software Defined Network

Software-defined networking (SDN) has transformed network management by decoupling the control and data planes, thereby enabling enhanced flexibility, programmability, and automation in network management. However, this flexibility introduces significant challenges to access control, particularly at the Northbound Interface (NBI), where external applications interact with SDN controller. Traditional access control mechanisms, such as rule-based frameworks and Role-Based Access Control (RBAC), suffer from inefficiencies, high maintenance overhead, and limited adaptability to evolving security policies. This paper proposes a novel access control framework that automates policy interpretation and management in SDN controller. The framework uses Natural Language Processing (NLP) techniques. It converts human-readable security policies into rules that machines can execute. This reduces the need for manual intervention and makes the system more adaptable. The experimental results demonstrate 97% accuracy, a precision of 1.00, and a reduced error rate of 0.03, significantly outperforming traditional methods, which exhibit a lower F1-score of 0.24 and higher latency. These findings underscore NLP’s potential for automating and improving SDN access control, offering an efficient and adaptive access control solution. This study highlights the trade-off between processing time and accuracy and suggests further optimizations to reduce computational overhead while maintaining high security and performance. This study also discusses implementation challenges, resource trade-offs and practical integration steps required for real deployments.

Text 122402.pdf - Published Version Restricted to Repository staff only Download (2MB) Official URL or Download Paper: https://link.springer.com/article/10.1007/s00607-0...

Actions (login required)

View Item