An Enhanced Quality of Service-Based Trust Model for Cloud Computing

Firdhous, Mohamed; Budiarto, Rahmat

doi:10.1007/978-981-16-6890-6_39

Cited by 5 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By means of such third parties, stakeholders can cover certain peak workloads by hiring or purchasing on-demand services and resources, e.g., computing or network resources. Thereby, billions of business relations across operators belonging to different domains are conceived in the foreseeable future, in which the selection of trustworthy third parties is a capital decision [2].…”

Section: Introductionmentioning

confidence: 99%

SLA-Driven Trust and Reputation Management Framework for 5G Distributed Service Marketplaces

Valero

Theodorou

Pérez

et al. 2024

IEEE Trans. Dependable and Secure Comput.

View full text Add to dashboard Cite

The fifth generation (5G) of mobile telecommunications is characterized by massive growth in the number of stakeholders, interconnected devices, and available services distributed under different administrative domains. Distributed marketplaces aim at facilitating stakeholders in the quest and hiring of third party resources and services. Establishing trustworthiness in such an open ecosystem is a cornerstone for the final deployment of these marketplaces in 5G networks and beyond. Hence, building trust management systems that ensure the selection of reliable parties or assets in 5G distributed marketplaces is essential. Thus, a reputation-based trust management framework is proposed to analyze stakeholder behavior patterns and predict trust scores to establish trustworthy relationships across domains. Furthermore, an Service Level Agreement (SLA)-driven reward and punishment mechanism is designed and developed on top of the reputationbased trust framework. Such a mechanism enables continuously adapting trust scores by gathering breach predictions, breach detections, and SLA violations in real time. Furthermore, an edgebased use case is presented to contextualize our reputation-based framework in a tangible enforcement scenario. In conclusion, three experiments were conducted on real-life testbeds demonstrating that our framework fairly distinguishes bad-mouthing attacks with 67% accuracy, when 50% recommenders are corrupted, and is resilient to continuous misbehavior bursts.

show abstract

Section: Introductionmentioning

confidence: 99%