Fifth-generation (5 G) network slicing enables multiple logical networks to share the same infrastructure, each designed to support distinct Service Level Agreements (SLAs). While network slicing is well supported in the Radio Access Network (RAN) and 5 G Core (5GC), the Transport Network (TN) backhaul is often treated as a slice-agnostic forwarding substrate. As a result, resource contention on shared links can significantly degrade network performance, such as latency and throughput, breaking end-to-end SLAs. This paper addresses this gap by making the TN sliceaware through a lightweight framework that translates slice requirements into concrete backhaul policies. The framework (i) associates each slice with TN identifiers at domain boundaries, (ii) turns slice requirements into per-slice forwarding and scheduling policies (e.g., priority and queue selection, bandwidth bounds, and optional congestion signals), and (iii) exports perslice measurements for monitoring and closed-loop control. We deploy a proof of concept on a programmable switch and evaluate it on a real-life 5 G testbed. Our results show that, under link saturation, slices designed for mission-critical services preserve their SLAs, while best-effort traffic degrades predictably.

Teleoperation is increasingly used to complement Level-4 autonomous driving, but it depends on strict network requirements i.e., latency below 5 ms, uplink throughput exceeding 25 Mbps, and reliability of 99.999%. To meet such requirements Network Slicing has emerged as a key paradigm to allocate and isolate network resources. However, in current 5G Standalone (5G SA)networks, Network Slicing faces fragmented orchestration. 3GPP mechanisms govern the 5G Core (5GC), O-RAN manages Radio Access Network (RAN) resources, while Transport Network (TN) operates separately, limiting end-to-end Quality of Service (QoS) guarantees.In this work, we propose a Cross-Domain Controller (CDC) that establishes synergy between network domains i.e., 5GC, TN, and RAN, through policy-driven coordination. The CDC implements a closed-loop, threshold-based mechanism that monitors slice Key Performance Indicators (KPIs) and triggers runtime network resource reconfigurations when the Teleoperation slice requires additional capacity to meet its QoS targets. Experimental validation on a real-world 5G SA testbed demonstrates that the CDC maintains strict slice isolation while dynamically expanding allocated network resources.

With the stringent requirements of evolving vertical services, there is a growing need for more flexible, ultra low-latency Beyond 5G (B5G)/6G architectures. One way to achieve this is by disaggregating core network functions and deploying them closer to the end users. The distribution and optimal placement of both Control Plane (CP) and User Plane (UP) functions at the network edge, along with the collocated orchestrated Application Functions (AFs), can significantly improve End-to-End (E2E) latency and network reliability by avoiding unnecessary traffic flows to the centralized cloud servers. Furthermore, as 5G evolves into 6G, automated and intelligent network management solutions will be needed in the distributed communication compute continuum. This paper presents an early-stage PhD research direction that focuses on the optimal quality-aware disaggregation of 5G core functions, aiming to reduce E2E latency and improve the throughput of 6G vertical services. The focus is on the optimal User Plane Function (UPF) placement, as well as the other CP functions and AFs, which altogether interact with the UPF. As UPF is directly involved in user data forwarding, it is crucial for handling user traffic over 5G/B5G network and, as such, it significantly impacts E2E latency. This paper provides i) an overview of theoretical concepts and State of the Art (SotA) UPF placement methodologies, and ii) future directions for optimized placement of B5G Core and application functions within the edge cloud continuum, leveraging intelligent network and service orchestration solutions.

We present LITE, a lightweight trajectory-unaware CSI estimation framework implemented as an O-RAN xApp and integrated within a CF-MaMIMO emulator. LITE compresses high-dimensional CSI at the O-DU, transports a compact latent representation over the midhaul, and predicts short-horizon channel gains at the Near-RT RIC using a compact SE-BiLSTM model. The demo highlights end-to-end real-time operation, interactive visualization, fault injection, and fallback mechanisms under realistic impairments such as missing or delayed measurements. Results show stable and accurate per-AP predictions while meeting Near-RT latency constraints, demonstrating the feasibility of embedding bandwidth-aware intelligence in O-RAN-compliant RAN loops.

Future 6G edge-intelligent radios require neural networks that satisfy strict latency and energy constraints, yet the practical behaviour of quantization-aware Neural Architecture Search (NAS) models on real Field-Programmable Gate Array (FPGA) hardware remains insufficiently explored. This work evaluates a set of Pareto-optimal architectures generated by a quantization-aware NAS MONAS-LQ, together with state-of-the-art reference models, when deployed on an FPGA using the Brevitas-FINN-Vivado toolchain. The results show that all models found by MONAS-LQ maintain accuracy within 0.30% of server-side execution, with several exhibiting slight improvements, while energy consumption remains below 20 mJ/sample and latency is dominated by early convolutional layers rather than overall model size. Compared to existing quantized architectures, the MONAS-LQ models achieve more favourable accuracy–efficiency trade-offs. These findings highlight the relevance of hardware-aware NAS for deriving deployable and energy-efficient Deep Learning (DL) models tailored to the resource constraints of future edge-intelligent radio systems.

Network Digital Twins (NDTs) enable safe what-if analysis for 6G cloud-edge infrastructures, but adoption is often limited by fragmented workflows from telemetry to validation. We present a data-driven NDT framework that extends 6G-TWIN with a scalable pipeline for cloud-edge telemetry aggregation and semantic alignment into unified data models. Our contributions include: (i) scalable cloud-edge telemetry collection, (ii) regime-aware feature engineering capturing the network's scaling behavior, and (iii) a validation methodology based on Sign Agreement and Directional Sensitivity. Evaluated on a Kubernetes-managed cluster, the framework extrapolates performance to unseen high-load regimes. Results show both Deep Neural Network (DNN) and XGBoost achieve high regression accuracy (R2>0.99), while the XGBoost model delivers superior directional reliability (Sa>0.90), making the NDT a trustworthy tool for proactive resource scaling in out-of-distribution scenarios.

Cell-Free Massive Multiple-Input Multiple-Output (CF-MaMIMO) in Open Radio Access Network (O-RAN) promises high spectral efficiency but is limited by frequent Channel State Information (CSI) exchanges, which strain fronthaul/midhaul/backhaul (X-haul) bandwidth and exceed the capabilities of existing approaches relying on uncompressed CSI or heavy predictors. To overcome these constraints, we propose LITE, a lightweight pipeline combining a 1-D convolutional Autoencoder (AE) at the O-RAN Distributed Unit (O-DU) with a Squeeze-and-Excitation (SE)-enhanced Bidirectional Long Short-Term Memory (BiLSTM) predictor at the Near-Real-Time RAN Intelligent Controller (Near-RT-RIC), enabling short-horizon trajectory-unaware forecasting under strict transport and processing budgets. LITE applies 50 % CSI compression and an asymmetric SE-BiLSTM, reducing model complexity by 83.39 % while improving accuracy by 5 % relative to a baseline BiLSTM. With compression-aware training, the Lightweight Intelligent Trajectory Estimator (LITE) incurs only 6 % accuracy loss versus the BiLSTM baseline, outperforming independent and end-to-end strategies. A TensorRT-optimized implementation achieves $147 k$ Queries per Second (QPS), a 4.6x throughput gain. These results demonstrate that LITE delivers X-haul-efficient, low-latency, and deployment-ready channel-gain prediction compatible with O-RAN splits.

Despite major advances in Connected and Autonomous Vehicles (CAVs), edge cases such as unmapped construction zones or dense urban areas still require human intervention. Teleoperation, often combined with Level 4 automation, addresses these situations but demands strict network performance i.e., latency below 5 ms, uplink throughput above 25 Mbps, and 99.999% reliability, to ensure safe, responsive control under heavy load. This paper presents a Proof-of-Concept (PoC) 5G Standalone (SA) network supporting end-to-end Network Slicing for teleoperation. The PoC achieves seamless slice isolation across the 5G Core (5GC), Transport Network (TN), and Radio Access Network (RAN), enabling dynamic and cross-domain Network Slicing to meet teleoperation requirements.

This demonstration paper presents a real-life 5G Standalone proof-of-concept showcasing end-to-end network slicing with both inter-slice and intra-slice isolation across the 5G Core, Transport Network, and Radio Access Network domains. The system, based on a smart hospital scenario, highlights a decentralized 5G architecture that aligns slice selection, controlplane, and data-plane mechanisms to guarantee that network resources are securely partitioned across the entire infrastructure. The demonstration highlights the technical implementation of slice orchestration and performance isolation, showing how critical services such as robotic-surgery control or vital-sign sensors maintain guaranteed quality of service even under high load, while less critical services coexist without interference. The proof-of-concept confirms the practical viability of complete end-to-end network slicing for mission-critical applications that demand reliability, confidentiality, and predictable performance.

This paper presents a comprehensive benchmark study of a real-life 5G Standalone (SA) deployment with different Multiple Input Multiple Output (MIMO) configurations (1x1, 2x2, and 4x4) in an indoor office environment. We evaluate the impact of distance, obstacles, and material composition on key performance metrics, including throughput, Reference Signal Received Power (RSRP), Signal-to-Interference-plus-Noise Ratio (SINR), and Rank Indicator (RI). The results demonstrate that while higher-order MIMO configurations can deliver substantial throughput gains under favorable conditions, their effectiveness is fundamentally constrained by environmental factors such as signal attenuation, multipath propagation, and material-induced losses. Our results provide practical guidelines for indoor network planning and optimization, establishing concrete performance baseline for open-source 5G systems in typical office scenarios highlighting the critical importance of site selection.

The exponential growth in connectivity and computing demand has made Network Function Virtualization Infrastructure (NFVI) a major contributor to global energy consumption. Conventional network and service deployments, whether based on legacy hardware appliances or NFVI stacks, struggle to dynamically provision resources for peak data traffic demand. This results in nearly constant energy consumption, even during low-traffic periods, leading to inefficient resource use. Network softwarization and virtualization have enabled flexible and programmable service deployments, which are beneficial for the rapid and dynamic scaling of network functions. This paper validates energy-aware service and network orchestration with a Zero-touch Network and Service Management (ZSM) framework for the autonomous optimization of computing, network, and power resources from NFVI in a use case focused on connected mobility, and in particular, smart traffic management. By modeling road traffic based on vehicle count and type and 3rd Generation Partnership Project (3GPP) profiles for data formats in vehicular communication scenarios, the ZSM framework adjusts services and resources to service requirements and to actual demand. Experimental validation on the real-life Smart Highway testbed in Antwerp, Belgium, demonstrates a strong correlation between vehicular traffic and power consumption, supporting the hypothesis that adaptive compute and network resource management reduces unnecessary energy use and advances the vision of sustainable and self-optimizing Sixth-Generation (6G) networks.

The distribution of services across heterogeneous edge and cloud infrastructures in Beyond 5G (B5G) and Sixth-Generation (6G) networks increases operational complexity, while existing MANagement and Orchestration (MANO) solutions remain single-domain and lack unified telemetry or autonomous cross-domain decision mechanisms. Prior work proposes cross-domain orchestration concepts, but most validations rely on simulation and overlook interoperability issues, non-stationary latency, and inconsistent Key Performance Indicator (KPI) models in real Network Function Virtualization Infrastructure (NFVI) deployments. This paper introduces a modular, technology-agnostic Zero-touch Network and Service Management (ZSM) framework that provides unified NFVI abstraction and a multi-criteria decision engine that jointly considers latency, compute load, and energy consumption for autonomous service placement, scaling, and dynamic resource provisioning. Evaluation in a real-world deployment across geographically distributed testbeds demonstrates that our ZSM framework mitigates latency spikes by up to 96%, reduces overloaded-node latency by 36%, and maintains stable performance under variable load. These results confirm the practicality and effectiveness of zero-touch, multi-domain orchestration for future 6G compute-continuum environments.

Network slicing enables multiple virtual networks to share physical resources, allowing network operators to deliver highly customizable and efficient networking solutions that meet the diverse requirements of modern applications. Automating the deployment and runtime management of network slices is essential for supporting network scaling with minimal human intervention. With several applications with different performance requirements, the manual configuration of slices is infeasible. Different initiatives have investigated methods in which network operators define high-level Key Performance Indicators that a network must deliver to networked applications, while the details of how to achieve these goals are abstracted by underlying control systems. Many works in the literature present a top-down approach, focusing on the high-level decision processes and relying on abstracted infrastructure managers and simulation tools to apply/execute such decisions. In this work, we leverage components that we previously developed for network monitoring, traffic shaping, and Software-Defined Time-Sensitive Networking to create a bottom-up approach toward automated slice management driven by Quality of Service goals. We describe the intricate coordination of elements required for an automated control loop and present the results achieved with a proof of concept executed in a real testbed of wired and Wi-Fi nodes. The applications use the data models introduced in this work to inform their requirements to the control plane. The control plane correctly prepares the network by applying rules for monitoring, traffic classification, and traffic shaping, effectively creating slices with different performance profiles for each application. Lastly, we present results using our control loop that continuously monitors each flow, identifies issues affecting the network performance, and takes corrective actions to reestablish the intended performance level of the applications.

Internet of Things (IoT) devices are increasingly being deployed in critical applications, such as eHealth systems, enabled by advancements in 5G technology, which offer more than 100 Mbps of throughput, less than 5 ms of latency, and $99,999 \%$ of reliability. However, to overcome computing limitations and security measures, IoT devices rely on cloudbased solutions to outsource data processing. This dependency introduces significant security concerns, as sensitive data must be transmitted over the network and processed in external environments, increasing the risk of interception, unauthorized access, and data breaches. To mitigate these security risks, within the scope of the MOZAIK project, we deploy Network Slicing to ensure end-to-end inter-slice and intra-slice isolation across all network domains i.e., 5G Core (5GC), Transport Network (TN), and Radio Access Network (RAN). We deploy a synergy across the entire network infrastructure i.e., $5 \mathrm{GC}, \mathrm{TN}$, and RAN, to isolate the IoT data flows from the moment the data is generated until it reaches the cloud, safeguarding sensitive data during transmission. The results of our real-life experiments demonstrate that our proof of concept provides robust isolation between slices, effectively addressing the security concerns of IoT devices and enhancing the reliability and security of IoT applications. Additionally, we also include aspects of secure data storage and secure data processing, covered in the MOZAIK project.