A two-way trust routing scheme to improve security in fog computing environment

A two-way trust routing scheme to improve security in fog computing environment

연구 분야: Safety

학회: Cluster Computing

Compliance with security requirements in the fog computing environment is known as an important phenomenon in maintaining the quality of service due to the dynamic topology. Security and privacy breaches can occur in fog computing because of its properties and the adaptability of its deployment method. These characteristics render current systems inappropriate for fog computing, including support for high mobility, a dynamic environment, geographic distribution, awareness of location, closeness to end users, and absence of redundancy. Although efficient secure routing protocols have been developed by researchers in recent years, it is challenging to ensure security, reliability, and quality of service at the same time to overcome the limitations of cloud-fog computing. In light of the fact that trust management is an effective means of protecting sensitive information, this study proposes a two-way trust management system (TMS) that would enable both the service requester and the service provider to verify each other's reliability and safety. The trustworthiness of the service seeker can also be verified in this way. So that fog clients can confirm that fog nodes can deliver suitable, dependable, and secure services, trust in a fog computing environment should ideally be two-way. The ability to verify the authenticity of fog clients is an important capability for fog nodes to have. A distributed, event-based, multi-trust trust system is presented by the suggested approach to trust computation, which makes use of social relationships (nodes and clients) and service quality criteria. Hence, the trust score is computed using a number of characteristics. Here, the weight of direct and indirect ratings is emphasized, and the final trust score is computed by dynamically merging the information gained from self-observation and the suggestions of nearby nodes. An extensive evaluation of the proposed method shows that it is resistant to a large number of badly behaved nodes and can successfully neutralize trust-based attacks.