Proactive caching in cellular networks effectively reduces backhaul load and content delay. Open Radio Access Networks (O-RAN), with their multi-vendor, distributed architecture, offer additional opportunities to cache content at various components. Integrating proactive caching with O-RAN can significantly improve performance, though the process is complex due to the tight coupling between content placement and O-RAN elements. This paper proposes an Intelligent Content-Aware Proactive Caching (ICAPC) mechanism for both published and unpublished videos. In ICAPC, video features are extracted and compressed using a 3D-CNN to generate high-dimensional feature vectors. Videos with similar features are clustered into 'video classes' using k-means clustering. The feature vector is mapped into a G-dimensional space, where each value represents the percentage of features from each video class. A support vector machine (SVM) is then trained using these vectors and their corresponding video popularity. For unpublished videos, the trained SVM predicts popularity based on their feature representation. Videos are cached based on descending popularity-first at the RU, then DU, and finally CU. Simulation results show that the proposed method outperforms conventional approaches by achieving higher cache hit ratios and reducing both content latency and backhaul load.

The integration of modulation techniques with machine learning in optical wireless communication (OWC) provides a solution for optical wireless access networks (OWANs) that require high-speed and high-reliability wireless connections. An environment-aware differential modulation and detection (DMD) scheme based on a convolutional neural network (CNN) in optical wireless communication systems is proposed in this paper. The scheme adopts a differential modulation and detection method that can mitigate the BER floor limitation of on-off keying (OOK) modulation in atmospheric turbulence channels. In addition, the machine learning approach is proposed for optimization, which notably obviates the need for intricate channel state estimation. In a deep neural network, the detector can extract amplitude features from multi-received signals. Furthermore, an experimental platform is set up for sampling the fluctuation of light intensity. Based on the experiments, the results demonstrate that the scheme exhibits significant performance advantages and effectively improves the system performance of the traditional DMD method in the low SNR region. Performance of the CNN-based method is also in-depth analyzed and compared with other methods of modulation and detection under varying scintillation indices. The insights and investigations provide the probability for the practical application of machine learning in differential OWC system design.

Ultra-high-speed railway communication systems face several technical challenges due to extremely high mobility, including Doppler-induced channel variations, frequent handovers, and increasing network traffic. These challenges not only degrade communication reliability but also negatively affect the efficiency of network resource utilization. In this paper, we review the key technical challenges in ultra-high-speed railway communication environments and investigate artificial intelligence (AI)-based intelligent network control techniques to address these issues. In particular, we examine mobility management approaches focusing on AI-based predictive handover schemes and intelligent network control architectures based on the Open Radio Access Network (O-RAN). In addition, network resource management strategies are discussed through mobile edge computing (MEC)-enabled traffic offloading and task migration techniques. Through this analysis, we discuss the potential applicability of intelligent network control technologies for improving communication reliability and enhancing network resource utilization efficiency in ultra-high-speed railway communication environments.

Accurate short-horizon downlink throughput prediction is essential for automation in high-density 5G deployments (e.g., stadiums and events), where user load, scheduling decisions, and interference conditions change rapidly and produce highly variable user-perceived rates. This paper benchmarks lightweight regression models for per-user throughput prediction from readily available radio access network (RAN) key performance indicators (KPIs) and studies a risk-aware extension that augments point forecasts with calibrated uncertainty and an abstention (deferral) rule. Experiments use a strictly time-ordered train/calibration/test protocol on the Liverpool 5G High-Density Demand (L5GHDD) dataset. The target is strongly zero-inflated (about 62% of samples at 0 Mbps) and heavy-tailed, creating regimes where average-error optimization can mask rare but operationally important bursts. In the point-prediction benchmark, the best model is a tuned two-stage support vector regressor with a mean absolute error (MAE) of 0.452 Mbps, while the strongest single-stage model attains a weighted mean absolute percentage error (WMAPE) of 56.200%. For uncertainty quantification, we compare standard split conformal prediction against two input-adaptive alternatives. Constant-width split conformal attains 88.900% marginal coverage for a nominal 90% target with an average interval width of 2.288 Mbps, but width-based deferral is degenerate because all intervals have the same size. Variable-length conformal intervals preserve near-nominal coverage (91.100%) while producing informative width variation: normalized conformal reduces the average width to 1.344 Mbps, and conformalized quantile regression reduces it to 0.641 Mbps. At a deferral threshold of 1.5 Mbps, constant-width conformal defers all samples, whereas normalized conformal still acts on 61.200% of samples with selective MAE 0.219 Mbps. These results show that input-adaptive uncertainty is necessary for meaningful selective prediction in heteroscedastic 5G throughput dynamics.

The transition 4G to 5G and beyond in the era of mobile communication has posed unprecedented challenges on scalability, energy efficiency (EE) and sustainability within the Radio Access Network (RAN) segment.In this paper, we propose a renewable-aware convex optimization model and develop a framework applicable to Open Radio Access Network (ORAN) architectures addressing joint Virtual machines (VM) placement, bandwidth allocation, and power control under hybrid grid-renewable energy environments.The proposed model includes an explicit renewable smoothing mechanism in order to overcome the fluctuation of renewable power supply, all the while guaranteeing a good Quality of Service (QoS) and Quality of Experience (QoE) provisioning.Obtaining the problem as a convex optimization ensures global optimality and computational feasibility in large-scale applications too.A formal rounding-and-repair algorithm is proposed to convert relaxed solutions into deployable discrete schedules with bounded integrality gap.Extensive simulations show that the approach can realize a well-balanced resource provisioning between grid and renewables, which in turn provides substantial EE improvement and smooth powerprofile transformation among time slots.Comprehensive comparisons against (Mixed-Integer Problem) MILP, greedy heuristic, and deep reinforcement learning (DRL) baselines demonstrate that the proposed approach achieves within 2.1% of optimal while being 47 faster than MILP.A comparison between the solver types shows that SQP gives the fastest convergence with fewest iterations and same good solution for real-time ORAN optimization.Results illustrate that the smoothing weight parameter () is an appropriate policy control knob, providing a direct balance between renewable stability and total operational cost.Systematic parameter sweeps and stress tests validate robustness under varying network scales, realistic renewable traces, and binding advanced constraints including interference management and QoE provisioning.In summary, the research lays an in-depth groundwork for long-term and sustainable ORAN systems with renewable energy supply to support future carbon-neutral 6G varieties with smart power management and adaptive resource optimization.

Mobile Edge Computing (MEC) has become increasingly critical for latency-sensitive applications, including Augmented/Extended Reality (AR/XR), Cloud Gaming, Real-Time Video Analytics, and Interactive Enterprise Services. Existing edge steering mechanisms remain largely reactive by relying on static policies, nearest-edge selection, or compute-only information that usually fail under user mobility, fluctuating radio conditions, dynamic user-plane paths, and edge resource contention rather than being more proactive. This paper presents SCONE-AEGIS framework that extends the Standard Communication with Network Elements (SCONE) paradigm beyond throughput advisories to support joint network-compute steering of MEC applications. SCONE-AEGIS introduces an Edge Steering Advice (ESA) that communicates recommendations that can be consumed by the applications, which have been derived from a combination of RAN, UPF, and MEC telemetry. The framework is a combination of a two-stage AI/ML engine, the first being a Spatio-Temporal Graph Predictor that is uncertaintyaware and models the evolving relationships among radio access network nodes, user-plane functions, edge sites, and mobile users, and the second stage is a Safe Contextual Bandit Steering Policy (SCBSP) that selects execution sites subject to SLA constraints, migration hysteresis, and prediction confidence. The proposed framework provides a standards-compatible, privacypreserving path for exposing joint network-compute intelligence to applications without breaking transport encryption.

Traditional radio access networks (RANs) are often characterized by rigid, proprietary architectures, high capital (CAPEX) and operational (OPEX) expenditures, and vendor lock-in. These limitations hinder cost-effective deployments in remote and underserved regions. In this paper, we evaluate the potential of Open RAN, a flexible and disaggregated networking paradigm leveraging virtualization and multi-vendor interoperability, to bridge the healthcare gap in Brazil through remote eHealth applications. Focusing on the mobile specialist practice (MSP) use case, we introduce a total cost of ownership model that integrates both CAPEX and OPEX across urban, suburban, and rural deployment scenarios, thereby addressing a critical gap in current network economic models. Additionally, we explore a detailed taxonomy of MSP sub-use cases that delineates the technical requirements for high-definition video streaming, haptic feedback, and remote diagnostics. We conducted laboratory experiments on a 5G Open RAN testbed using real ultrasound equipment in collaboration with Samsung, Beneficência Portuguesa (one of the largest private healthcare hubs in Latin America), and InovaHC (the innovation center of the region’s largest hospital). These experiments demonstrated that our approach meets stringent performance indicators such as low latency, robust throughput, and reliable packet delivery under diverse backhaul conditions. The results further indicate that Open RAN can substantially reduce deployment costs while ensuring high-performance remote medical services. These findings provide valuable insights for advancing field trials, refining economic models, and guiding future policy initiatives.

Fully decoupled radio access networks (FD-RANs) achieve spectral efficiency and coverage flexibility for 6G via independent uplink (UL) and downlink (DL) base station operation, yet dynamic user mobility brings critical challenges to joint user association and resource allocation. Asymmetric interference and heterogeneous base station capacities cause persistent network unfairness, while uncoordinated mobility management triggers ping-pong handovers and heavy handover overheads. To resolve these intertwined problems, we propose a fully decoupled, mobility-resilient and fairness-guaranteed framework, which integrates short-term congestion pricing with the long-term Jain fairness index for equitable resource distribution and introduces a composite handover penalty with a strict physical hysteresis margin to block invalid handovers. We formulate the optimization problem as a novel Sliding-Window Hysteresis-Integrated Fairness Two-Layer Multi-Armed Bandit (SHIFT-MAB) model, embedding an exponentially weighted moving average (EWMA) sliding-window mechanism to track real-time channel fluctuations efficiently. Theoretical analysis confirms the model's decoupling optimality, sublinear regret bound and fairness convergence. Extensive simulations show that SHIFT-MAB effectively suppresses invalid handovers, ensures high network fairness, optimizes system utility and achieves a superior handover-throughput trade-off.

Future radio access networks must accommodate growing mobile data traffic. Capacity between mobile devices and remote radio units (RRUs) must increase by using higher wireless carrier frequencies. Higher frequencies reduce reach, requiring RRU densification and high-capacity front- and backhaul connections. Optical fiber offers high throughput, but deployment can be expensive or unfeasible. Here we show an all-photonic sub-THz wireless link at 226 GHz over 1400 m, achieving a record-high net-rate-distance product of 214.2 Gbit s−1 km. Broadband photonic and plasmonic components enabled flat frequency response at high speeds. A novel dual-sideband receiver increased signal-to-noise ratio by 2 dB. We assessed power variations due to atmospheric turbulence; the scintillation index remained below 0.019 under strong turbulence, confirming sub-THz link resilience. A theoretical comparison with free-space optical links highlights the turbulence resistance. In this work, we show sub-THz links offer high capacity, resilience to weather and turbulence, and cost-effective deployment for wireless front- and backhaul.

Future 6G technology needs to meet dynamic requirements in diverse scenarios, as recommended by the International Telecommunication Union (ITU). The conventional monolithic radio access network (RAN) equipped with dedicated hardware fails to meet the rapid iteration and dynamic requirements of networks. Consequently, there is a need to reform RAN architecture and management strategies. The paper proposes and demonstrates a dynamic service-based RAN. The flexibility of the RAN is enhanced by decoupling network capabilities, including both functional and state decoupling. RAN comprises services that can be independently deployed. These services exhibit stateless characteristics, enabling them to be registered and discovered by the system. Furthermore, the Data Plane and Digital Plane are introduced to efficiently manage highly flexible services, thereby establishing a functional network. Services can be effectively managed to facilitate network orchestration and scaling through on-demand scheduling. The current research examines the differences between static and dynamic networks in their approaches to handling evolving scenarios. By dynamically scaling services to meet user requirements, the network avoids unnecessary scaling of all functions at the RAN level. The dynamic network has been demonstrated to be more efficient and cost-effective in scenarios where the number of users and data rates fluctuate.

Microgrid for radio access network resilience against power grid outages: Design and operation

It is well-known that, without access to application-layer parameters controlled by Over-The-Top (OTT) providers, Mobile Network Operators (MNOs) struggle to accurately predict customers’ Quality of Experience (QoE). While some previous proposals have suggested interaction between OTTs and MNOs, they have faced challenges in terms of practical implementation and limited application scenarios. This work aims to advance these solutions with two key contributions. First, following the Open Radio Access Network (O-RAN) architecture, we propose adding components that integrate a machine learning (ML)-based QoE prediction model, deployed by the MNO, into the O-RAN system. By establishing specific data-sharing interfaces between OTTs and MNOs, our approach helps MNOs overcome the limitations in updating their quality prediction modules. Second, we present a network-aware, ML-driven QoE prediction model that captures the relationship between the resulting QoE and various network parameters, such as signal-to-interference-noise ratio (SINR), channel quality indicator (CQI), network resource blocks (RBs), throughput, and device mobility. Among seven considered ML regressors, the Gradient Boosting (GB) achieved the highest QoE prediction performance in terms of R 2 (0.906) and RMSE (0.259).

Optimizing relay selection and wireless resource allocation in Proximity Radio Access Networks

The introduction of a new Apple device (iPhone, iPad, Apple Watch) into global markets requires structured integration with diverse carrier ecosystems. Carrier needs to setup new product which is getting launched with various device capabilities attributes, compatible SIM profiles (SA or NSA), supply chain related information etc. for product order, provisioning and activation. Since operators are transitioning to advanced 5G deployments—including Standalone (SA), Non-Standalone (NSA) as well as some MVNOs enabling CBRS architectures—device launch depends on concurrent certification across radio access networks (RAN), core networks, IMS subsystems, roaming frameworks, and handovers with MNO-CBRS, WIFI Networks. This paper presents a structured view of how Apple device launches intersect with MVNO carrier integration, focusing on operational launch activities (carrier settings readiness, activation testing, retail/on-device flows), and business considerations (distribution, support, compliance, and lifecycle management). It covers architectural alignment, certification plans, interoperability testing, risk governance, and post-launch analytics. A case study illustrates integration challenges and mitigation strategies. The proposed lifecycle model enables swift rollout of New Apple devices while maintaining network stability, compliance, and premium user experience.

The high-paced development of wireless communication technologies brought about the growing need of smart, autonomous, and ultra-high-performance network structures, and the combination of Artificial Intelligence and Machine Learning in the 6G wireless communication networks is a major research issue. The conventional network management techniques cannot support the new demands that include ultra-low latency communications, intelligent spectrum management, mass connectivity, and real-time dynamic resource provision. This paper is a systematically-conducted literature review on the recent research on AI-driven networks, edge AI, federated learning, intelligent radio access network design, and AI-native architecture in next-generation networks based on PRISMA. The review focuses on the possibilities of self-organizing networks, digital twin networks, and smart connectivity that machine learning, deep learning, and autonomous wireless systems provide to the next generation of wireless networks working in terahertz communication ranges and integrated sensing and communication landscape. The findings indicate that AI based resource allocation, intelligent surfaces of reconfigurability, massive mimo optimization and network automation are all effective towards improving spectral efficiency, reliability and energy efficiency of 6G wireless communication. Moreover, the paper has also noted the increased importance of smart security mechanisms, distributed learning, and edge computing in enabling scalable and secure AI-native architectures. The results also indicate that the next generation wireless networks will be based upon intelligent spectrum management, autonomous control, and real-time data analytics to accommodate the new applications in holographic communication, smart cities, extended reality, and connected autonomous systems.

ABSTRACT Open Radio Access Network (O‐RAN) fronthaul links face stringent bandwidth, latency, and computational constraints, which become particularly critical when transmitting high‐resolution medical images. This paper proposes an adaptive region‐aware image compression framework for healthcare imaging over O‐RAN that reduces fronthaul load while preserving diagnostically relevant information. Each image is partitioned into Region‐of‐Interest (ROI) and Non‐ROI areas and compressed using independent quantisation parameters. An optimisation model is formulated to minimise transmitted data size subject to ROI and Non‐ROI quality constraints, end‐to‐end latency bounds and computational limits at O‐RAN nodes. The framework is evaluated using two medical imaging datasets (chest X‐rays and bone fracture X‐rays), where empirical rate–distortion and quality models are derived and validated. Results demonstrate substantial fronthaul bandwidth reduction—achieving compression ratios up to 416:1—while maintaining ROI quality and diagnostic accuracy above 97%. These findings highlight the effectiveness of region‐aware optimisation for bandwidth‐efficient healthcare imaging in O‐RAN environments.

The rapid growth of wireless communication systems and devices has increased global energy consumption in wireless communication networks. Therefore, wireless communication technologies are advancing towards the development of fifth-generation wireless communication systems and beyond, i.e., towards the development of sixth-generation wireless communication systems. It is important that the energy efficiency of wireless communication systems considered a significant parameter along with data rate, delay and reliability. Green wireless communication systems aim to reduce the power consumption of wireless communication systems minimize carbon emissions and make wireless communication systems environmentally sustainable. This paper presents a comprehensive review of sustainable wireless communication systems and energy-efficient protocols for green wireless communication systems. This study discusses various architectural strategies such as energy-efficient base stations, heterogeneous and ultra-dense networks, cloud radio access networks, adaptive networks, renewable energy and their role enables wireless networks to modify their behavior according to the nature of the data and availability of resources. Further, the paper discusses the role of sustainable wireless protocols in all layers of wireless communication, where techniques like power control, duty cycle, energy-efficient routing and transport protocols are playing a significant role in removing the unnecessary communication and utilizing resources efficiently. It also discusses enabling technologies that can be utilized in energy-efficient network operation including energy harvesting, cognitive radio, artificial intelligence, edge computing and network virtualization. Moreover, some of the major challenges in energy-efficient wireless communication including decisions, complexity, security, cost, and regulatory issues, are addressed. This study provides an sustainable wireless communication can be achieved based on recent developments and existing research gaps in wireless communication particularly in future 5G and 6G networks.

Abstract This article presents an experimentation environment, empirical study and results of runtime energy consumption in 5th generation (5G) networks and discusses the potential solutions to enhance energy efficiency with radio access network (RAN) and application control. Further, the paper evaluates the actual power consumption of selected end user devices and applications in 5G RAN covering indoor small-cell installations and 5G outdoor macro sites. The evaluation setup utilized a state-of-the-art 5G radio access network (RAN) alongside efficient, accurate energy measurement equipment, measuring power consumption in live scenarios with selected video streams at varying bit rates to reflect advanced video streaming configurations. The outcome of the research shows that several enhancements, such as lower total energy consumption and longer end device battery life, are possible when advanced control is applied to the existing and future wireless mobile networks. The experimental results provide solid power consumption figures and lead to detailed reasons for the holistic view of energy enhancements covering the overall end-to-end data path including the end user devices, RAN as well as edge and cloud infrastructure and applications—addressing these aspects at a scale not covered in earlier studies. The increasing demand for high-quality digital services, particularly video streaming, presents significant challenges for reducing energy consumption in mobile networks; as these networks and services continue to grow, this study contributes to ongoing efforts to minimize their runtime energy consumption.

Open radio access network (RAN) leverages the RAN intelligent controller (RIC) to enable artificial intelligence/machine learning (AI/ML)-driven network automation. However, a gap remains between algorithmic research and deployable, standards-compliant rApp prototypes with verifiable behavior. This paper addresses this gap by introducing an integrated development and validation framework that supports the full lifecycle of AI/ML-based rApps, from prototyping to functional verification. The framework includes a standards-compliant non-real-time RIC (Non-RT RIC) architecture with supporting functions, an interface for integrating RAN simulators, and a visualization dashboard that displays system state and control actions, enabling traceability of end-to-end control loops. We demonstrate the framework through a case study involving the design and implementation of a predictive network energy saving rApp. In closed-loop experiments, instrumented logs and visualizations indicate that the control decisions of the rApp adhere to the intended operational logic, allowing repeatable functional validation. We also discuss challenges for real-world deployment and study limitations. Overall, the proposed framework provides a practical methodology and toolset that accelerate the transition from algorithmic concept to deployable, validated rApps, advancing reliable AI/ML solutions within the O-RAN ecosystem and offering direct applicability to energy saving as well as other O-RAN use cases.

ABSTRACT The next‐generation wireless technology modernizes the telecommunication infrastructure by incorporating cloud‐based services into its existing framework. Cloud radio access network (C‐RAN) was introduced in the 4G era and became popular in recent trends due to the flexibility of sharing resources and decoupling of the data plane and control plane. While this decoupling improves radio resource allocation and processing efficiency, it also introduces challenges in optimizing routes between user equipment (UE), remote radio heads (RRHs), and baseband units (BBUs) under diverse QoS requirements. In this work, the route optimization problem for 5G C‐RAN architecture is formulated by developing an optimization function for a dense urban network that consists of 500 UEs connected to 100 RRHs and further linked to 20 BBUs via fronthaul connections. To enhance path selection and routing optimization, several nature‐inspired algorithms, that is, ACO, WOA, PSO, GWO, CSO, BOA, and SMO are applied, as these methods are effective for solving complex optimization problems. The performance of these algorithms is analyzed in terms of best fitness value, network efficiency, number of satisfied users, number of blocked users, and number of dropped users under high traffic density. To reduce the impact of randomness in the search process, the experimental results are statistically validated using ANOVA and Tukey's post hoc tests. A detailed comparison of nature‐inspired algorithms utilized in this work has been carried out to evaluate their performance, which further confirms the practical suitability and advantages of ACO in 5G C‐RAN systems.