The rise of Software-Defined Networking (SDN) has revolutionized network management, offering greater flexibility and programmability. However, ensuring the accuracy of packet forwarding remains paramount for maintaining network reliability and security in SDN environme
More
The rise of Software-Defined Networking (SDN) has revolutionized network management, offering greater flexibility and programmability. However, ensuring the accuracy of packet forwarding remains paramount for maintaining network reliability and security in SDN environments. Unlike traditional IP networks, SDN separates the control plane from the data plane, creating new challenges for securing data transmission. Existing verification methods designed for IP networks often cannot be directly applied to SDN due to this architectural difference. To address the limitations of existing verification methods in SDN networks, new approaches are necessary. This research proposes a novel parallel method for verifying packet forwarding, building upon concepts from DYNAPFV. The proposed approach aims to overcome specific limitations of existing methods (including DYNAPFV), such as scalability issues, slow verification times. Simulations demonstrate significant improvements compared to DYNAPFV. The proposed parallel method achieves a 92% reduction in time required to identify malicious nodes within the network. The results also reveal a trade-off between security and verification time. As the probability of packet integrity confirmation increases from 0.8 to 0.99, system security strengthens, but the time to detect malicious switches also increases.
Manuscript profile
