Edge-Native Multi-Agent System For Latency-Critical Anomaly Detection And Qos Optimization In 6g Smart Manufacturing

Author

• Ndidi Nzeako Anyakora, Prairie View A&M University

Date of Award

8-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Degree Discipline

Electrical Engineering

Abstract

The increasing reliance on wireless connectivity in smart manufacturing places stringent demands on network latency, reliability, and adaptability that exceed the capabilities of static or threshold-based quality-of-service (QoS) control mechanisms. While 5G standalone networks support industrial deployments, emerging 6G environments are expected to introduce greater variability, tighter latency constraints, and increased performance degradation, necessitating autonomous, learning-driven control strategies.

This dissertation proposed an intelligent multi-agent framework for real-time anomaly detection and QoS optimization in industrial wireless networks by integrating edge-based monitoring, unsupervised anomaly detection, and reinforcement learning-driven control. Network telemetry collected from a Firecell 5G standalone testbed served as an empirical reference, while a scaled 6G-emulated dataset evaluated the framework's robustness under more stringent performance conditions. Isolation Forest and deep autoencoder models identified deviations in latency, throughput, and packet loss. QoS control was formulated as a Markov Decision Process and addressed using reinforcement learning agents that dynamically adjust network parameters to mitigate detected anomalies and maintain desired network performance.

The proposed multi-agent system was evaluated against static and threshold-based reference methods across multiple operating regimes. Experimental results demonstrated improved latency stability, faster throughput recovery, and reduced packet loss under moderate and stress-intensive network conditions. The learning-driven framework maintained consistent control behavior across 5G-and 6G-oriented scenarios, demonstrating robustness under increased network volatility. By coordinating monitoring, anomaly detection, and adaptive QoS control at the network edge, the framework enabled proactive responses to network degradation. These findings validated the effectiveness of multi-agent, learning-driven QoS optimization and provided a practical pathway toward autonomous network management in future 6G-enabled smart manufacturing environments.

Index Terms: 6G smart manufacturing, edge-native anomaly detection, multi-agent reinforcement learning (MARL), quality of service (QoS) optimization, ultra-reliable low-latency communication (URLLC).

Committee Chair/Advisor

Cajetan Akujuobi

Committee Member

Suxia Cui

Committee Member

Mohamed Chouikha

Committee Member

Justin Foreman

Committee Member

Noushin Ghaffari

Publisher

Prairie View A&M University

Rights

Date of Digitization

8/27/2026

Contributing Institution

J. B Coleman Library

City of Publication

Prairie View

MIME Type

Application/PDF

Recommended Citation

Anyakora, N. (2026). Edge-Native Multi-Agent System For Latency-Critical Anomaly Detection And Qos Optimization In 6g Smart Manufacturing. Retrieved from https://digitalcommons.pvamu.edu/pvamu-dissertations/143