Seokhoon Kim

The transition toward Open Radio Access Network (O-RAN) architecture has enabled unprecedented intelligence and flexibility in 5G and 6G network slicing. However, a fundamental challenge remains in managing the tension between radio unit energy efficiency and the strict Service Level Agreement (SLA) requirements of Ultra-Reliable Low-Latency Communication (URLLC) slices, particularly under highly dynamic traffic conditions. Existing O-RAN approaches suffer from a timescale conflict where Non-Real-Time (Non-RT) policy planners optimize for long-term energy but fail to react to rapid traffic surges, while Near-Real-Time (Near-RT) controllers prioritize reliability at the cost of significant energy over-provisioning. To address this, we propose H-RLS, a hierarchical multi-timescale framework that decouples control into a Non-RT Proximal Policy Optimization (PPO) agent for strategic, energy-aware policy planning and a Near-RT Recursive Least Squares (RLS)-assisted xApp. By predicting millisecond-level delay risks, the xApp acts as a mathematically constrained safety net, applying bounded tactical adjustments when critical SLA violations are detected. Extensive evaluations across dynamic traffic transitions demonstrate that H-RLS maintains zero SLA violations. By actively preventing resource over-provisioning, the framework achieves the lowest composite Energy-SLA cost across all tested regimes, significantly minimizing dynamic power consumption while preserving Enhanced Mobile Broadband (eMBB) service integrity.

Multi-access edge computing (MEC) enables computation-intensive perception and decision-making tasks in vehicular networks to be offloaded to nearby edge servers. Existing approaches usually fix the artificial intelligence (AI) inference model, overlooking how model selection jointly affects latency, energy consumption, and service reliability. We propose a hierarchical model selection and control (HMSC) framework based on deep reinforcement learning (DRL) for MEC-assisted vehicular networks. The framework couples a vehicle-layer MAPPO component that provides a communication interface representation for subchannel assignment and energy accounting with a centralized MEC-layer soft actor-critic (SAC) agent that, under SDN orchestration, adaptively selects lightweight or high-fidelity AI models and allocates computational resources. Accordingly, the core contribution of this paper lies in MEC-side model-aware computation control under an explicitly defined subchannel-contention abstraction, rather than in physical-layer transmit-power optimization. Both layers are guided by a composite objective that integrates normalized end-to-end (E2E) latency, normalized energy consumption, and a deadline-violation penalty. Using a discrete-time simulation framework, HMSC reduces E2E latency compared with static inference and non-hierarchical DRL baselines and sustains a higher deadline satisfaction ratio (DSR) under constrained uplink throughput and varying traffic loads. The learned policy is load-aware, favoring high-fidelity inference under light load and lightweight inference under congestion; a post hoc analysis using YOLOv5-family accuracy reference further quantifies the inference-quality implications of this adaptive selection behavior. These results show that coordinated MEC-side control of AI model selection and computation, under a shared deadline-aware objective, provides a robust latency–energy trade-off for MEC-assisted vehicular 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.

Federated Learning (FL) has been proposed as a new machine learning paradigm to ensure data privacy by training the model in a decentralized manner. However, FL is challenged by device heterogeneity, asymmetric data contribution, and imbalanced datasets, which complicate system control and hinder performance due to long waiting times for aggregation. To tackle the FL challenges, we propose Adaptive Time Threshold Client Selection using DRL (ATCS-FL) to adjust the time threshold (α) in each communication round based on computing and resource capacity of each device and the volume of data updates. The Double Deep Q-Network (DDQN) model determines the appropriate α, according to the variations in local training time that achieves performance improvement alongside latency reduction. Based on the α, the server selects a subset of clients with adequate resources that can finish training within the α for participating in the training process. Our approach dynamically adjusts the α and adaptively selects the number of clients, effectively mitigates the impact of heterogeneous training speeds and significantly enhances communication efficiency. Our experiment utilizes CIFAR-10 and MNIST benchmarked datasets for image classification training with convolutional neural networks across non-IID distributed levels in FL. Specifically, ATCS-FL demonstrates performance improvement and latency reduction of 77% and 75%, respectively, compared to FedProx and FLASH-RL.

The rapid increase in the deployment of Internet of Things (IoT) sensor networks has led to an exponential growth in data generation and an unprecedented demand for efficient resource management infrastructure. Ensuring end-to-end communication across multiple heterogeneous network domains is crucial to maintaining Quality of Service (QoS) requirements, such as low latency and high computational capacity, for IoT applications. However, limited computing resources at multi-access edge computing (MEC), coupled with increasing IoT network requests during task offloading, often lead to network congestion, service latency, and inefficient resource utilization, degrading overall system performance. This paper proposes an intelligent task offloading and resource orchestration framework to address these challenges, thereby optimizing energy consumption, computational cost, network congestion, and service latency in dynamic IoT-MEC environments. The framework introduces task offloading and a dynamic resource orchestration strategy, where task offloading to the MEC server ensures an efficient distribution of computation workloads. The dynamic resource orchestration process, Service Function Chaining (SFC) for Virtual Network Functions (VNFs) placement, and routing path determination optimize service execution across the network. To achieve adaptive and intelligent decision-making, the proposed approach leverages Deep Reinforcement Learning (DRL) to dynamically allocate resources and offload task execution, thereby improving overall system efficiency and addressing the optimal policy in edge computing. Deep Q-network (DQN), which is leveraged to learn an optimal network resource adjustment policy and task offloading, ensures flexible adaptation in SFC deployment evaluations. The simulation result demonstrates that the DRL-based scheme significantly outperforms the reference scheme in terms of cumulative reward, reduced service latency, lowered energy consumption, and improved delivery and throughput.

Recently, Network Functions Virtualization (NFV) has become a critical resource for optimizing capability utilization in the 5G/B5G era. NFV decomposes the network resource paradigm, demonstrating the efficient utilization of Network Functions (NFs) to enable configurable service priorities and resource demands. Telecommunications Service Providers (TSPs) face challenges in network utilization, as the vast amounts of data generated by the Internet of Things (IoT) overwhelm existing infrastructures. IoT applications, which generate massive volumes of diverse data and require real-time communication, contribute to bottlenecks and congestion. In this context, Multi-access Edge Computing (MEC) is employed to support resource and priority-aware IoT applications by implementing Virtual Network Function (VNF) sequences within Service Function Chaining (SFC). This paper proposes the use of Deep Reinforcement Learning (DRL) combined with Graph Neural Networks (GNN) to enhance network processing, performance, and resource pooling capabilities. GNN facilitates feature extraction through Message-Passing Neural Network (MPNN) mechanisms. Together with DRL, Deep Q-Networks (DQN) are utilized to dynamically allocate resources based on IoT network priorities and demands. Our focus is on minimizing delay times for VNF instance execution, ensuring effective resource placement, and allocation in SFC deployments, offering flexibility to adapt to real-time changes in priority and workload. Simulation results demonstrate that our proposed scheme outperforms reference models in terms of reward, delay, delivery, service drop ratios, and average completion ratios, proving its potential for IoT applications.

The last decade has witnessed the explosive growth of the internet of things (IoT), demonstrating the utilization of ubiquitous sensing and computation services. Hence, the industrial IoT (IIoT) is integrated into IoT devices. IIoT is concerned with the limitation of computation and battery life. Therefore, mobile edge computing (MEC) is a paradigm that enables the proliferation of resource computing and reduces network communication latency to realize the IIoT perspective. Furthermore, an open radio access network (O-RAN) is a new architecture that adopts a MEC server to offer a provisioning framework to address energy efficiency and reduce the congestion window of IIoT. However, dynamic resource computation and continuity of task generation by IIoT lead to challenges in management and orchestration (MANO) and energy efficiency. In this article, we aim to investigate the dynamic and priority of resource management on demand. Additionally, to minimize the long-term average delay and computation resource-intensive tasks, the Markov decision problem (MDP) is conducted to solve this problem. Hence, deep reinforcement learning (DRL) is conducted to address the optimal handling policy for MEC-enabled O-RAN architectures. In this study, MDP-assisted deep q-network-based priority/demanding resource management, namely DQG-PD, has been investigated in optimizing resource management. The DQG-PD algorithm aims to solve resource management and energy efficiency in IIoT devices, which demonstrates that exploiting the deep Q-network (DQN) jointly optimizes computation and resource utilization of energy for each service request. Hence, DQN is divided into online and target networks to better adapt to a dynamic IIoT environment. Finally, our experiment shows that our work can outperform reference schemes in terms of resources, cost, energy, reliability, and average service completion ratio.

AI-empowered applications have been deployed in many aspects of networking, and federated learning (FL) has emerged as a complementary approach due to its ability to enable privacy-preserving model training and inference. However, without self-organizing capability, practical FL systems face several issues to co-exist in real-time networking. Therefore, this paper aims to design autonomous FL management with integrated graph neural networks (GNN) and deep reinforcement learning (DRL), namely AutoFedGDRL, to sustain heterogeneous FL execution in optimized open radio access network (O-RAN) and intelligent core network architectures and offer automated policy-driven orchestration by intelligent agent controller. Edge cloud virtualized O-RAN is integrated to assist model computation and support multiple services with elastic containerized resource scaling. The practicability of FL systems is stimulated by modelling the participants and aggregators as a graph representation and subsequently analyzing to predict the accessibility and trustworthiness of the nodes, bandwidth capacities, and virtual link relationship. Our proposed AutoFedGDRL aims to obtain specifications of hidden FL, service, and networking states in order to control the main policies, such as training management, resource sharing, aggregation scheduling, and service prioritization. In the experiment, AutoFedGDRL surpassed reference models (non-federated training) in global accuracy, achieving 98.23% for MNIST and 97.12% for CIFAR-10, compared to 98.22% and 95.89% for PrimaryGNN-FL. The proposed scheme also improved end-to-end convergence speed, with execution times 10.58 ms to 32.79 ms faster. Model delivery ratios reached 99.98%, ensuring the federated system's reliability and sharing workload efficiency.