Alawadh, Rehab, Yadav, Poonam orcid.org/0000-0003-0169-0704 and Ahmadi, Hamed orcid.org/0000-0001-5508-8757 (Accepted: 2025) DYNAPARC: AI-Driven Predictive Path Failure Management for Industrial IoT-Fog Networks. In: DYNAPARC: AI-Driven Predictive Path Failure Management for Industrial IoT-Fog Networks. IEEE Communications Society (In Press)
Abstract
The increasing adoption of IoT-Fog networks in industrial environments demands resilient systems to meet stringent Quality-of-Service (QoS) requirements. Network failures disrupt critical processes and degrade QoS, necessitating innovative predictive failure management. This paper presents the Dynamic Resilient Path Recovery (DYNAPARC) system, an AI-centric solution leveraging Software-Defined Networking (SDN) to predict and mitigate failures in industrial IoT-Fog networks (IIoT). DYNAPARC integrates AI-based reliability prediction model with SDN's programmable architecture and routing protocols to enhance resilience. A hybrid approach combines proactive and reactive methods: secondary paths are pre-installed (proactively) for immediate failover during primary link failures, while new alternative paths are dynamically calculated in real-time (reactively) for multiple failures, ensuring adaptive routing. To quantify the system’s performance, a novel Network Performance Score (N) measures QoS under failure conditions. Simulations show that DYNAPARC maintains an N score above 0.975135 before and after failures, outperforming traditional reactive and proactive methods. Integrating machine learning in the SDN controller significantly reduces packet loss by selecting the most reliable paths. These results highlight the potential of AI-driven prediction and SDN to achieve predictive reliability, ensuring superior resilience, fast recovery, and efficient traffic management in fog-based IIoT environments.
Metadata
Item Type: | Proceedings Paper |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy. |
Dates: |
|
Institution: | The University of York |
Academic Units: | The University of York > Faculty of Sciences (York) > Computer Science (York) The University of York > Faculty of Sciences (York) > Electronic Engineering (York) |
Funding Information: | Funder Grant number EPSRC EP/Y019229/1 EPSRC EP/X040518/1 |
Depositing User: | Pure (York) |
Date Deposited: | 25 Feb 2025 14:50 |
Last Modified: | 25 Feb 2025 14:50 |
Status: | In Press |
Publisher: | IEEE Communications Society |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:223728 |
Download
