2026· IEEE Transactions on Network and Service Management· Vol 23, pp. 6523-6535· 0 citations· 46 references
Abstract
As Software-Defined Networking (SDN) and Network Function Virtualization (NFV) enabled networks scale in size and complexity, monitoring and managing Service Function Chains (SFCs) under stringent latency and resource constraints becomes increasingly challenging. Although Deep Reinforcement Learning (DRL) is widely applied to SFC provisioning and Virtual Network Function (VNF) placement, enhanced network state monitoring is crucial to capture unexpected network conditions and guide DRL agents toward more adaptive decisions. In this context, Language Models (LMs) enable flexible, natural-language (NL)–based, query-driven network monitoring; however, directly processing complex multi-metric NL queries is computationally expensive and error-prone. This paper proposes an end-to-end (E2E) edge-based query translation pipeline that decomposes multi-metric NL queries into simpler single-metric sub-queries. Query decomposition is performed using a retrieval-augmented language model (RAG-LLM) and compared with a lightweight rule-based decomposition baseline. The resulting sub-queries are translated into Structured Query Language (SQL) using FLAN-T5. A cloud-only baseline, which directly translates NL queries to SQL without decomposition, is also evaluated. The results show that the rule-based edge pipeline achieves the lowest latency, reducing E2E latency by up to 78% compared to RAG-LLM and 18% compared to cloud execution under high workloads. Under increasing arrival rates for the largest workload, the rule-based edge pipeline maintains superior performance over cloud, reducing total E2E latency by 57% at $\lambda = 0.8$ . While RAG-LLM provides greater flexibility for unseen query patterns, both edge-based approaches achieve 100% NL2SQL accuracy with zero decomposition failures, outperforming the cloud-only baseline (95% accuracy).
Efficient Service Function Chaining (SFC) orchestration in large-scale NFV environments is often bottlenecked by the non-linear overhead of traditional Graph Convolutional Networks (GCNs). This paper proposes GraphSAGE-DQN, a scalable algorithm that leverages inductive neighbor sampling to decouple state extraction complexity from network size, achieving linear computational complexity. By integrating GraphSAGE embeddings with a Deep Q-Network (DQN), the model optimizes deployment strategies to maximize request acceptance rates. Simulations show that GraphSAGE-DQN outperforms traditional DQN and SECA algorithms, particularly in high-load scenarios where it improves acceptance rates by $\mathbf{1. 6 4 \%}$ and $\mathbf{1 7. 0 \%}$, respectively. These results confirm the model's efficiency and scalability for dynamic SFC orchestration in large-scale networks.
Network functions virtualization (NFV) is an emerging technology that enables flexible service deployment for supporting the Beyond 5G/6G network. NFV transforms physical network devices into virtual network functions (VNF) over Edge Computing capabilities, thereby facilitating the agility of network services and reducing management costs. To effectively monitor Internet of Things (IoT) network resources, service function chaining (SFC) is used for its virtualizations to ensure the multi-service requirements are sufficiently in capability, scalability, and flexibility for computation workloads alignments. However, to satisfy the resource availability requirements and efficiency under several conditions, SFC reconfiguration methods face the challenges in meeting significant latency requirement of delay-sensitive applications while reaching the importance of energy saving on orchestration timespan. In this paper, we propose task management-aware SFC and orchestrating schemes, namely GNN-PPO. In this framework, we utilize the Graph Neural Network (GNN), which relies on the message-passing neural network (MPNN), to capture all the abstraction of physical resource nodes and link capabilities over MEC node states. In particularly, GNN is divided construction into two phrases: (1) GNN represents nodes for all the Mobile edge computing (MEC) nodes, which have a global view on resources of computation and communicational capabilities that could serve as carriers; (2) VNFs are transferred into graph networks by using feature-extraction MPNN to manage each VIM that seeks an optimal and reliable analysis of traffic fluctuations. Lastly, Deep Reinforcement Learning (DRL) is used to embrace the network determination in policy strategy, which utilizes a Proximal Policy Gradient (PPO). On the other hand, we propose a novel network architecture based on PPO to perform the design for the optimization of resource utilization and facilitate energy consumption on MEC servers under diverse setting scenarios, which enables continuous policy enforcement for our system. With the experimental results, we compare our proposed solution with reference schemes in terms of rewards with learning rate and batch size, average request acceptance, SFC success, packet delivery, throughput, and resource utilization ratio that confirm the scheme’s scalability and practical suitability for IoT network deployment.
Seyha Ros, Taikuong Iv, Intae Ryoo et al.· Italian National Conference...· 0 citations
Efficient long-term network evolution is becoming increasingly critical in dense 5G-Advanced and beyond cellular systems, where persistent traffic imbalances and localized congestion pose significant challenges that conventional short-term radio resource management alone cannot fully mitigate. This paper proposes a digital twin (DT)-enabled non-real-time (NRT) network evolution framework integrated with a large language model (LLM). Within this architecture, the digital twin provides a high-fidelity, controllable environment for evaluating infrastructure actions, while the LLM serves as a strategic orchestration engine that recommends cost-efficient network upgrades based on observed network states. Unlike traditional optimization methods that require exhaustive mathematical reformulations for each specific scenario, the proposed framework leverages the reasoning capabilities of LLMs to interpret operator objectives and constraints in natural language, generating structured evolution plans. The considered NRT action space encompasses antenna upgrades, bandwidth expansion, and new base station (BS) deployment. A techno-economic formulation is introduced to jointly evaluate load reduction performance and overall economic expenditure. Numerical results in a dense cellular scenario demonstrate that the framework effectively reduces peak resource utilization and provides diverse, coordinated evolution strategies tailored to varying network conditions.
Yukai Wang, Janghee Woo, G. Hahm et al.· International Conference on...· 0 citations
Fifth generation (5G) networks deliver multi-gigabit data rates, sub-millisecond latency, and dense connectivity through a customised service delivery paradigm built on virtualisation and network slicing (NS). However, conventional NS frameworks rely on threshold-based control and lack the context awareness needed for autonomous, user-centric decision-making. Machine learning (ML) optimization-driven methods such as deep reinforcement learning (DRL) and hybrid metaheuristic–ML approaches can close this gap by inferring user bandwidth behaviour, anticipating congestion, and enacting proactive corrective actions. This paper presents a systematic, PRISMA-based review of 2024–2026 ML-based optimization for user-centric distributed NS, screening 6,286 records to 67 core studies that are normalised through a common evidence tuple. A comprehensive critical review is then presented with role-oriented matrices spanning admission control, resource allocation and offloading, orchestration, graph learning, and federated learning (FL). From this synthesis, we derive recurring optimization formulations and identify persistent gaps, namely the absence of direct Quality of Experience (QoE) inference, privacy preservation, topology awareness, and validated deployment. To address these gaps, we propose a QoE-aware framework for Multi-Access Edge Computing (MEC)-enabled Open Radio Access Network (O-RAN) architectures, combining a graph attention network (GAT) encoder, distributed multi-agent DRL, and privacy-preserving FL, while transitioning control from Quality of Service (QoS) to QoE metrics. The proposed framework is also grounded in preliminary validation from our two prior slice admission control and load balancing studies, offering a scalable, privacy-aware, and truly user-centric path towards 5G and Beyond 5G networks.
Manoj Prasad Kunasegran, Wai Leong Pang, S. K. Phang· IEEE Access· 0 citations
A constrained optimization model that supports different management goals through alternative objective functions (latency-aware or power-aware) while enforcing operational constraints, including node capacities, slice-specific latency bounds, and explicit limits on VNF migrations/relocations between scheduling periods is proposed.
R. Moreno-Vozmediano, E. Huedo, R. Montero et al.· Journal of Network and Syste...· 0 citations
The complexity of modern network infrastructures continues to increase, demanding integrated evaluation across multiple layers—from physical-layer signal propagation to TCP/IP protocol performance. Multi-level network simulators have become indispensable for such analysis, offering cost-effective and risk-free experimentation platforms. However, leveraging these simulators remains challenging due to steep learning curves and extensive domain-specific knowledge requirements. To address this challenge, we introduce a network-oriented Large Language Model (LLM) that serves as an intelligent intermediary between users and simulators, enabling interactive, automated, and script-free simulation workflows. Using NVIDIA Sionna and ns-3 as representative case studies, we adapt a general-purpose LLM into a domain-specialized model through parameter-efficient fine-tuning and retrieval-augmented generation (RAG). The resulting agentic LLM integrates accurate physical-layer modeling with automated higher-layer evaluation, executing complete multi-layer workflows via natural language queries. This framework not only simplifies complex configuration and orchestration tasks but also significantly reduces the barrier to simulation-driven research and education. Our results demonstrate that such network-oriented LLMs can generalize across simulators and network layers, paving the way for a new paradigm of AI-assisted, multi-level network experimentation and design.
Dongming Wu, Jiewen Liu, Xingqin Lin et al.· IEEE Transactions on Network...· 0 citations