A Centralized Reputation Management Scheme for Isolating Malicious Controller(s) in Distributed Software-Defined Networks

Abstract

Abstract-Software-Defined Networks have seen an increasing in their deployment because they offer better network manageability compared to traditional networks. Despite their immense success and popularity, various security issues in SDN remain open problems for research. Particularly, the problem of securing the controllers in distributed environment is still short of any solutions. This paper proposes a scheme to identify any rogue/malicious controller(s) in a distributed environment. Our scheme is based on trust and reputation system which is centrally managed. As such, our scheme identifies any controllers acting maliciously by comparing the state of installed flows/policies with policies that should be installed. Controllers rate each other on this basis and report the results to a central entity, which reports it to the network administrator.

Improvement of User Information Behavior in Online Medical Community Based on Reputation Incentive Game

“…A punishment mechanism is used to enhance the cooperative behavior of nodes in wireless ad hoc networks [9]. A trust driven model for network architecture software is proposed [10], which enables trust management and online evolution of trust relationship to be realized. Based on the law of trust, it aims to accumulate feedback of different information interactions to calculate the reputation of users, and then dynamically adjust the reputation of users through reputation rewards and punishments for different information behaviors.…”