TrustBlock: An adaptive trust evaluation of SDN network nodes based on double-layer blockchain

Area of Science:

• Computer Science
• Network Security

Background:

• Software Defined Networks (SDN) rely on data layer devices for efficient packet forwarding.
• The efficiency and security of node forwarding tasks in SDN have been underexplored.

Purpose of the Study:

• To propose TrustBlock, a novel method integrating trust attributes into SDN routing.
• To enhance the integrity and controllability of trust evaluation in SDN using a double-layer blockchain architecture.

Main Methods:

• A double-layer blockchain architecture records node behavior data and performs trust calculations.
• Trust is evaluated based on direct, indirect, and historical trust metrics.
• The entropy method is employed for adaptive weighting of evaluation attributes.

Main Results:

• The TrustBlock model achieved a detection rate of up to 98.89% for abnormal nodes in SDN.
• Blockchain ensures node identity authentication and reliability through forwarding status analysis.
• Consensus algorithms and adaptive historical trust weights mitigate malicious recommendations and periodic attacks.

Conclusions:

• TrustBlock effectively identifies abnormal nodes in SDN, enhancing network security.
• The proposed double-layer blockchain architecture offers integrity and controllability in trust evaluation.
• This approach addresses the need for efficient and secure packet forwarding in SDN environments.