The proliferation of connected objects in the Internet of Things (IoT) ecosystem presents challenges in enabling real-time intelligence while safeguarding data privacy, especially within the computational and energy limitations of edge devices. This study, based on simulation and synthetic data collection methods, addresses these challenges by proposing a lightweight edge AI framework that employs federated learning, enabling model training across distributed Internet of People and Things (IoP) devices without transferring raw data to centralised servers. The framework integrates on-device inference, stochastic local updates, and model compression to ensure low-latency decision-making while adhering to memory and energy constraints. To enhance security, differential privacy mechanisms, encrypted aggregation, and robust outlier detection are utilized to defend against adversarial and Byzantine attacks. The proposed framework offers an effective solution for deploying federated intelligence on resource-constrained IoP devices, enabling responsive, privacy-preserving, and resilient edge AI operations in distributed environments.
F. Philip-Kpae, A. Imoize, Lloyd Endurance Ogbondamati et al.· E3S Web of Conferences· 0 citations
5G networks increasingly rely on key enabling technologies such as Software-Defined Networking (SDN), Network Function Virtualization (NFV), Multi-Access Edge Computing (MEC), and end-to-end network slicing to deliver heterogeneous services with strict quality-of-service (QoS) guarantees. However, programmability, multi-tenancy, and distributed edge–cloud operation significantly expand the attack surface. At the same time, traditional rule-based and reactive security mechanisms remain slow to adapt and may violate latency constraints during mitigation. This paper addresses the problem of QoS-compliant, closed-loop security control for sliced SDN/NFV infrastructures. We propose an AI-assisted, cross-layer security orchestration framework that integrates epoch-wise telemetry with ML-based risk estimation and formalizes mitigation as a Constrained Markov Decision Process (CMDP). The CMDP controller selects enforceable actions— slice isolation, rate limiting, traffic rerouting, and key reconfiguration—while explicitly satisfying latency/overhead constraints, and executes them via SDN flow-rule updates and NFV policy/VNF reconfiguration. Simulation results over 50 decision epochs demonstrate effective response to an injected high-risk event: risk spikes to 0.95 at epoch 15, after which the controller drives risk toward ≈0.10 while maintaining latency below the 40ms QoS bound (with a transient rise during mitigation and subsequent stabilization). The reward trajectory briefly degrades during disruption but recovers and converges to a positive long-term return, indicating stable constraint-aware operation. This work provides (i) a deployable cross-layer orchestration architecture for sliced networks, (ii) a QoS-constrained CMDP decision model that converts risk signals into actionable SDN/NFV controls, and (iii) empirical evidence that adaptive mitigation can reduce security risk without sacrificing service guarantees.
F. Philip-Kpae, A. Imoize, K. C. Okafor et al.· E3S Web of Conferences· 0 citations